{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,17]],"date-time":"2026-10-17T16:54:18Z","timestamp":1792256058414,"version":"4.4.0"},"reference-count":196,"publisher":"Elsevier BV","license":[{"start":{"date-parts":[[2026,6,1]],"date-time":"2026-06-01T00:00:00Z","timestamp":1780272000000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/www.elsevier.com\/tdm\/userlicense\/1.0\/"},{"start":{"date-parts":[[2026,6,1]],"date-time":"2026-06-01T00:00:00Z","timestamp":1780272000000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/www.elsevier.com\/legal\/tdmrep-license"},{"start":{"date-parts":[[2026,4,6]],"date-time":"2026-04-06T00:00:00Z","timestamp":1775433600000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/100012774","name":"Innovation Fund Denmark","doi-asserted-by":"publisher","award":["2105-00013A"],"award-info":[{"award-number":["2105-00013A"]}],"id":[{"id":"10.13039\/100012774","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["elsevier.com","sciencedirect.com"],"crossmark-restriction":true},"short-container-title":["Image and Vision Computing"],"published-print":{"date-parts":[[2026,6]]},"DOI":"10.1016\/j.imavis.2026.105980","type":"journal-article","created":{"date-parts":[[2026,4,6]],"date-time":"2026-04-06T16:44:41Z","timestamp":1775493881000},"page":"105980","update-policy":"https:\/\/doi.org\/10.1016\/elsevier_cm_policy","source":"Crossref","is-referenced-by-count":4,"special_numbering":"C","title":["A survey on dynamic neural networks: From computer vision to multi-modal sensor fusion"],"prefix":"10.1016","volume":"170","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-4626-1455","authenticated-orcid":false,"given":"Fabio","family":"Montello","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6430-7405","authenticated-orcid":false,"given":"Ronja","family":"G\u00fcldenring","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0881-8344","authenticated-orcid":false,"given":"Simone","family":"Scardapane","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3620-4123","authenticated-orcid":false,"given":"Lazaros","family":"Nalpantidis","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"78","reference":[{"key":"10.1016\/j.imavis.2026.105980_b1","series-title":"A survey of quantization methods for efficient neural network inference","author":"Gholami","year":"2021"},{"key":"10.1016\/j.imavis.2026.105980_b2","doi-asserted-by":"crossref","first-page":"10558","DOI":"10.1109\/TPAMI.2024.3447085","article-title":"A survey on deep neural network pruning: Taxonomy, comparison, analysis, and recommendations","volume":"46","author":"Cheng","year":"2024","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"10.1016\/j.imavis.2026.105980_b3","doi-asserted-by":"crossref","first-page":"1789","DOI":"10.1007\/s11263-021-01453-z","article-title":"Knowledge distillation: A survey","volume":"129","author":"Gou","year":"2021","journal-title":"Int. J. Comput. Vis."},{"key":"10.1016\/j.imavis.2026.105980_b4","series-title":"Neural architecture search: Insights from 1000 papers","author":"White","year":"2023"},{"key":"10.1016\/j.imavis.2026.105980_b5","doi-asserted-by":"crossref","first-page":"185","DOI":"10.1016\/j.neucom.2020.02.035","article-title":"Sparse low rank factorization for deep neural network compression","volume":"398","author":"Swaminathan","year":"2020","journal-title":"Neurocomputing"},{"key":"10.1016\/j.imavis.2026.105980_b6","doi-asserted-by":"crossref","first-page":"485","DOI":"10.1109\/JPROC.2020.2976475","article-title":"Model compression and hardware acceleration for neural networks: A comprehensive survey","volume":"108","author":"Deng","year":"2020","journal-title":"Proc. IEEE"},{"key":"10.1016\/j.imavis.2026.105980_b7","doi-asserted-by":"crossref","DOI":"10.1177\/1059712313484502","article-title":"Integrating visual perception and manipulation for autonomous learning of object representations","author":"Schiebener","year":"2013","journal-title":"Adapt. Behav."},{"key":"10.1016\/j.imavis.2026.105980_b8","doi-asserted-by":"crossref","first-page":"1221","DOI":"10.1167\/10.7.1221","article-title":"fMRI decoding of natural scene categories from line drawings","volume":"10","author":"Walther","year":"2010","journal-title":"J. Vis."},{"key":"10.1016\/j.imavis.2026.105980_b9","series-title":"Dynamic neural networks: A survey","author":"Han","year":"2021"},{"key":"10.1016\/j.imavis.2026.105980_b10","series-title":"Findings of the Association for Computational Linguistics: EACL 2023","first-page":"2370","article-title":"A survey on dynamic neural networks for natural language processing","author":"Xu","year":"2023"},{"key":"10.1016\/j.imavis.2026.105980_b11","series-title":"Proceedings of the 22nd International ACM Conference on Modeling, Analysis and Simulation of Wireless and Mobile Systems","first-page":"279","article-title":"SEE: Scheduling Early exit for mobile DNN inference during service outage","author":"Wang","year":"2019"},{"key":"10.1016\/j.imavis.2026.105980_b12","series-title":"Proceedings of the 59th ACM\/IEEE Design Automation Conference","first-page":"481","article-title":"EcoFusion: Energy-aware adaptive sensor fusion for efficient autonomous vehicle perception","author":"Malawade","year":"2022"},{"key":"10.1016\/j.imavis.2026.105980_b13","series-title":"2024 IEEE Globecom Workshops","first-page":"1","article-title":"Adaptive semantic token selection for AI-native goal-oriented communications","author":"Devoto","year":"2024"},{"key":"10.1016\/j.imavis.2026.105980_b14","unstructured":"E.R. Kandel, J.D. Koester, S.H. Mack, S.A. Siegelbaum, Principles of Neural Science, sixth ed."},{"key":"10.1016\/j.imavis.2026.105980_b15","doi-asserted-by":"crossref","first-page":"20","DOI":"10.1016\/0166-2236(92)90344-8","article-title":"Separate visual pathways for perception and action","volume":"15","author":"Goodale","year":"1992","journal-title":"Trends Neurosci."},{"key":"10.1016\/j.imavis.2026.105980_b16","doi-asserted-by":"crossref","DOI":"10.1371\/journal.pbio.1002348","article-title":"Neuropeptide-driven cross-modal plasticity following sensory loss in caenorhabditis elegans","volume":"14","author":"Rabinowitch","year":"2016","journal-title":"PLoS Biol."},{"key":"10.1016\/j.imavis.2026.105980_b17","series-title":"2017 29th Chinese Control and Decision Conference","first-page":"3853","article-title":"Sensor fusion: A review of methods and applications","author":"Fung","year":"2017"},{"key":"10.1016\/j.imavis.2026.105980_b18","doi-asserted-by":"crossref","first-page":"954","DOI":"10.1007\/s12559-020-09734-4","article-title":"Why should we add early exits to neural networks?","volume":"12","author":"Scardapane","year":"2020","journal-title":"Cogn. Comput."},{"key":"10.1016\/j.imavis.2026.105980_b19","first-page":"175","article-title":"Conditional computation in neural networks: Principles and research trends","volume":"18","author":"Scardapane","year":"2024","journal-title":"Intell. Artif."},{"key":"10.1016\/j.imavis.2026.105980_b20","series-title":"Advances in Neural Information Processing Systems","first-page":"19974","article-title":"Chasing sparsity in vision transformers: An end-to-end exploration","author":"Chen","year":"2021"},{"key":"10.1016\/j.imavis.2026.105980_b21","doi-asserted-by":"crossref","DOI":"10.1016\/j.neucom.2024.127468","article-title":"Sparsity in transformers: A systematic literature review","volume":"582","author":"Farina","year":"2024","journal-title":"Neurocomputing"},{"key":"10.1016\/j.imavis.2026.105980_b22","first-page":"90:1","article-title":"Split computing and early exiting for deep learning applications: Survey and research challenges","volume":"55","author":"Matsubara","year":"2022","journal-title":"ACM Comput. Surv."},{"key":"10.1016\/j.imavis.2026.105980_b23","series-title":"Proceedings of the 2016 Design, Automation & Test in Europe Conference & Exhibition","first-page":"475","article-title":"Conditional deep learning for energy-efficient and enhanced pattern recognition","author":"Panda","year":"2016"},{"key":"10.1016\/j.imavis.2026.105980_b24","series-title":"Proceedings of the 36th International Conference on Machine Learning","first-page":"3301","article-title":"Shallow-deep networks: understanding and mitigating network overthinking","author":"Kaya","year":"2019"},{"key":"10.1016\/j.imavis.2026.105980_b25","series-title":"2016 23rd International Conference on Pattern Recognition","first-page":"2464","article-title":"BranchyNet: Fast inference via early exiting from deep neural networks","author":"Teerapittayanon","year":"2016"},{"key":"10.1016\/j.imavis.2026.105980_b26","doi-asserted-by":"crossref","first-page":"1875","DOI":"10.1109\/TPAMI.2019.2959322","article-title":"Anytime recognition with routing convolutional networks","volume":"43","author":"Jie","year":"2021","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"10.1016\/j.imavis.2026.105980_b27","series-title":"Multi-scale dense networks for resource efficient image classification","author":"Huang","year":"2018"},{"key":"10.1016\/j.imavis.2026.105980_b28","series-title":"2020 IEEE\/CVF Conference on Computer Vision and Pattern Recognition","first-page":"2366","article-title":"Resolution adaptive networks for efficient inference","author":"Yang","year":"2020"},{"key":"10.1016\/j.imavis.2026.105980_b29","series-title":"ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing","first-page":"1","article-title":"CHASE: channel-wise and spatial attention for early exiting in image classification","author":"Addad","year":"2025"},{"key":"10.1016\/j.imavis.2026.105980_b30","series-title":"Advances in Neural Information Processing Systems","first-page":"11960","article-title":"Not all images are worth 16x16 words: Dynamic transformers for efficient image recognition","author":"Wang","year":"2021"},{"key":"10.1016\/j.imavis.2026.105980_b31","series-title":"2023 IEEE\/CVF International Conference on Computer Vision","first-page":"777","article-title":"Dynamic token pruning in plain vision transformers for semantic segmentation","author":"Tang","year":"2023"},{"key":"10.1016\/j.imavis.2026.105980_b32","series-title":"2023 IEEE\/CVF International Conference on Computer Vision","first-page":"5969","article-title":"Dynamic perceiver for efficient visual recognition","author":"Han","year":"2023"},{"key":"10.1016\/j.imavis.2026.105980_b33","series-title":"Adaptive computation with elastic input sequence","author":"Xue","year":"2023"},{"key":"10.1016\/j.imavis.2026.105980_b34","series-title":"ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing","first-page":"4167","article-title":"Differentiable branching in deep networks for fast inference","author":"Scardapane","year":"2020"},{"key":"10.1016\/j.imavis.2026.105980_b35","series-title":"Learning to stop while learning to predict","author":"Chen","year":"2020"},{"key":"10.1016\/j.imavis.2026.105980_b36","first-page":"9825","article-title":"Empowering adaptive early-exit inference with latency awareness","volume":"35","author":"Tan","year":"2021","journal-title":"Proc. AAAI Conf. Artif. Intell."},{"key":"10.1016\/j.imavis.2026.105980_b37","doi-asserted-by":"crossref","first-page":"1","DOI":"10.3390\/e24010001","article-title":"A probabilistic re-intepretation of confidence scores in multi-exit models","volume":"24","author":"Pomponi","year":"2022","journal-title":"Entropy"},{"key":"10.1016\/j.imavis.2026.105980_b38","series-title":"EERO: Early Exit with Reject Option for Efficient Classification with limited budget","author":"Valade","year":"2024"},{"key":"10.1016\/j.imavis.2026.105980_b39","doi-asserted-by":"crossref","first-page":"2263","DOI":"10.3390\/math12142263","article-title":"To exit or not to exit: Cost-effective early-exit architecture based on Markov decision process","volume":"12","author":"Kim","year":"2024","journal-title":"Mathematics"},{"key":"10.1016\/j.imavis.2026.105980_b40","series-title":"BEEM: Boosting performance of early exit DNNs using multi-exit classifiers as experts","author":"Bajpai","year":"2025"},{"key":"10.1016\/j.imavis.2026.105980_b41","series-title":"2019 IEEE International Workshop on Signal Processing Systems","first-page":"178","article-title":"DynExit: A dynamic early-exit strategy for deep residual networks","author":"Wang","year":"2019"},{"key":"10.1016\/j.imavis.2026.105980_b42","series-title":"2016 IEEE Conference on Computer Vision and Pattern Recognition","first-page":"770","article-title":"Deep residual learning for image recognition","author":"He","year":"2016"},{"key":"10.1016\/j.imavis.2026.105980_b43","doi-asserted-by":"crossref","DOI":"10.1016\/j.knosys.2023.111003","article-title":"Self-supervised efficient sample weighting for multi-exit networks","volume":"280","author":"Liu","year":"2023","journal-title":"Knowl.-Based Syst."},{"key":"10.1016\/j.imavis.2026.105980_b44","series-title":"2020 IEEE\/ACM International Conference on Computer Aided Design","first-page":"1","article-title":"HAPI: Hardware-Aware Progressive Inference","author":"Laskaridis","year":"2020"},{"key":"10.1016\/j.imavis.2026.105980_b45","series-title":"Multiple-exit tuning: Towards inference-efficient adaptation for vision transformer","author":"Liu","year":"2024"},{"key":"10.1016\/j.imavis.2026.105980_b46","series-title":"Proceedings of the Thirty-Seventh AAAI Conference on Artificial Intelligence and Thirty-Fifth Conference on Innovative Applications of Artificial Intelligence and Thirteenth Symposium on Educational Advances in Artificial Intelligence","first-page":"8657","article-title":"Predictive exit: Prediction of fine-grained early exits for computation-and energy-efficient inference","author":"Li","year":"2023"},{"key":"10.1016\/j.imavis.2026.105980_b47","doi-asserted-by":"crossref","DOI":"10.1016\/j.patcog.2020.107346","article-title":"Efficient adaptive inference for deep convolutional neural networks using hierarchical early exits","volume":"105","author":"Passalis","year":"2020","journal-title":"Pattern Recognit."},{"key":"10.1016\/j.imavis.2026.105980_b48","doi-asserted-by":"crossref","DOI":"10.1016\/j.cviu.2023.103900","article-title":"A closer look at branch classifiers of multi-exit architectures","volume":"239","author":"Lin","year":"2024","journal-title":"Comput. Vis. Image Underst."},{"key":"10.1016\/j.imavis.2026.105980_b49","series-title":"2019 IEEE\/CVF International Conference on Computer Vision","first-page":"1891","article-title":"Improved techniques for training adaptive deep networks","author":"Li","year":"2019"},{"key":"10.1016\/j.imavis.2026.105980_b50","series-title":"Improving the accuracy of early exits in multi-exit architectures via curriculum learning","author":"Bakhtiarnia","year":"2021"},{"key":"10.1016\/j.imavis.2026.105980_b51","series-title":"Computer Vision \u2013 ECCV 2022","first-page":"691","article-title":"Meta-GF: Training dynamic-depth neural networks harmoniously","volume":"vol. 13671","author":"Sun","year":"2022"},{"key":"10.1016\/j.imavis.2026.105980_b52","series-title":"Advances in Neural Information Processing Systems","first-page":"2516","article-title":"Zero time waste: Recycling predictions in early exit neural networks","author":"Wo\u0142czyk","year":"2021"},{"key":"10.1016\/j.imavis.2026.105980_b53","doi-asserted-by":"crossref","first-page":"580","DOI":"10.1016\/j.neunet.2023.10.003","article-title":"Zero time waste in pre-trained early exit neural networks","volume":"168","author":"W\u00f3jcik","year":"2023","journal-title":"Neural Netw."},{"key":"10.1016\/j.imavis.2026.105980_b54","series-title":"Computer Vision \u2013 ECCV 2022","first-page":"362","article-title":"Learning to weight samples for dynamic early-exiting networks","author":"Han","year":"2022"},{"key":"10.1016\/j.imavis.2026.105980_b55","series-title":"Boosted dynamic neural networks","author":"Yu","year":"2022"},{"key":"10.1016\/j.imavis.2026.105980_b56","series-title":"Jointly-learned exit and inference for a dynamic neural network: JEI-DNN","author":"Regol","year":"2024"},{"key":"10.1016\/j.imavis.2026.105980_b57","series-title":"2019 IEEE\/CVF International Conference on Computer Vision","first-page":"3712","article-title":"Be your own teacher: Improve the performance of convolutional neural networks via self distillation","author":"Zhang","year":"2019"},{"key":"10.1016\/j.imavis.2026.105980_b58","series-title":"2019 IEEE\/CVF International Conference on Computer Vision","first-page":"1355","article-title":"Distillation-based training for multi-exit architectures","author":"Phuong","year":"2019"},{"key":"10.1016\/j.imavis.2026.105980_b59","first-page":"10218","article-title":"Harmonized dense knowledge distillation training for multi-exit architectures","volume":"35","author":"Wang","year":"2021","journal-title":"Proc. AAAI Conf. Artif. Intell."},{"key":"10.1016\/j.imavis.2026.105980_b60","first-page":"4388","article-title":"Self-distillation: Towards efficient and compact neural networks","volume":"44","author":"Zhang","year":"2022","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"10.1016\/j.imavis.2026.105980_b61","series-title":"LGViT: Dynamic early exiting for accelerating vision transformer","author":"Xu","year":"2023"},{"key":"10.1016\/j.imavis.2026.105980_b62","doi-asserted-by":"crossref","first-page":"8043","DOI":"10.1007\/s40747-024-01572-3","article-title":"A multi-level collaborative self-distillation learning for improving adaptive inference efficiency","volume":"10","author":"Zhang","year":"2024","journal-title":"Complex Intell. Syst."},{"key":"10.1016\/j.imavis.2026.105980_b63","series-title":"2019 IEEE International Conference on Image Processing","first-page":"3048","article-title":"Adaptive inference using hierarchical convolutional bag-of-features for low-power embedded platforms","author":"Passalis","year":"2019"},{"key":"10.1016\/j.imavis.2026.105980_b64","series-title":"2021 29th European Signal Processing Conference","first-page":"1346","article-title":"Adaptive inference for face recognition leveraging deep metric learning-enabled early exits","author":"Passalis","year":"2021"},{"key":"10.1016\/j.imavis.2026.105980_b65","series-title":"Predicting probabilities of error to combine quantization and early exiting: QuEE","author":"Regol","year":"2024"},{"key":"10.1016\/j.imavis.2026.105980_b66","series-title":"2023 IEEE\/CVF International Conference on Computer Vision","first-page":"17322","article-title":"Window-based early-exit cascades for uncertainty estimation: When deep ensembles are more efficient than single models","author":"Xia","year":"2023"},{"key":"10.1016\/j.imavis.2026.105980_b67","series-title":"2024 IEEE\/CVF Winter Conference on Applications of Computer Vision","first-page":"2668","article-title":"Fixing overconfidence in dynamic neural networks","author":"Meronen","year":"2024"},{"key":"10.1016\/j.imavis.2026.105980_b68","series-title":"The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networks","author":"Frankle","year":"2019"},{"key":"10.1016\/j.imavis.2026.105980_b69","series-title":"2023 IEEE 31st Annual International Symposium on Field-Programmable Custom Computing Machines","first-page":"121","article-title":"ATHEENA: A toolflow for hardware early-exit network automation","author":"Biggs","year":"2023"},{"key":"10.1016\/j.imavis.2026.105980_b70","series-title":"2023 International Joint Conference on Neural Networks","first-page":"1","article-title":"EDANAS: Adaptive neural architecture search for early exit neural networks","author":"Gambella","year":"2023"},{"key":"10.1016\/j.imavis.2026.105980_b71","first-page":"1","article-title":"NACHOS: Neural architecture search for hardware-constrained early-exit neural networks","author":"Gambella","year":"2025","journal-title":"IEEE Trans. Neural Netw. Learn. Syst."},{"key":"10.1016\/j.imavis.2026.105980_b72","series-title":"Adaptive neural networks for efficient inference","author":"Bolukbasi","year":"2017"},{"key":"10.1016\/j.imavis.2026.105980_b73","series-title":"2020 IEEE\/ACM Symposium on Edge Computing","first-page":"84","article-title":"FlexDNN: Input-adaptive on-device deep learning for efficient mobile vision","author":"Fang","year":"2020"},{"key":"10.1016\/j.imavis.2026.105980_b74","series-title":"Proceedings of the 26th Annual International Conference on Mobile Computing and Networking","first-page":"1","article-title":"SPINN: Synergistic progressive inference of neural networks over device and cloud","author":"Laskaridis","year":"2020"},{"key":"10.1016\/j.imavis.2026.105980_b75","series-title":"Computer Vision \u2013 ECCV 2020","first-page":"275","article-title":"Early exit or not: Resource-efficient blind quality enhancement for compressed images","volume":"vol. 12361","author":"Xing","year":"2020"},{"key":"10.1016\/j.imavis.2026.105980_b76","doi-asserted-by":"crossref","unstructured":"A. Ghodrati, B.E. Bejnordi, A. Habibian, FrameExit: Conditional early exiting for efficient video recognition, in: Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, 2021, pp. 15608\u201315618.","DOI":"10.1109\/CVPR46437.2021.01535"},{"key":"10.1016\/j.imavis.2026.105980_b77","series-title":"Proceedings of the 30th ACM International Conference on Information & Knowledge Management","first-page":"823","article-title":"Dynamic early exit scheduling for deep neural network inference through contextual bandits","author":"Ju","year":"2021"},{"key":"10.1016\/j.imavis.2026.105980_b78","series-title":"Anytime dense prediction with confidence adaptivity","author":"Liu","year":"2021"},{"key":"10.1016\/j.imavis.2026.105980_b79","series-title":"ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing","first-page":"8488","article-title":"Branchy-GNN: A Device-edge co-inference framework for efficient point cloud processing","author":"Shao","year":"2021"},{"key":"10.1016\/j.imavis.2026.105980_b80","doi-asserted-by":"crossref","first-page":"461","DOI":"10.1016\/j.neunet.2022.06.038","article-title":"Single-layer vision transformers for more accurate early exits with less overhead","volume":"153","author":"Bakhtiarnia","year":"2022","journal-title":"Neural Netw."},{"key":"10.1016\/j.imavis.2026.105980_b81","series-title":"Computer Vision \u2013 ECCV 2022","first-page":"330","article-title":"Multi-exit semantic segmentation networks","volume":"vol. 13681","author":"Kouris","year":"2022"},{"key":"10.1016\/j.imavis.2026.105980_b82","series-title":"Temporal early exits for efficient video object detection","author":"Sabet","year":"2022"},{"key":"10.1016\/j.imavis.2026.105980_b83","series-title":"SEENN: Towards temporal spiking early-exit neural networks","author":"Li","year":"2023"},{"key":"10.1016\/j.imavis.2026.105980_b84","series-title":"2023 IEEE\/CVF Conference on Computer Vision and Pattern Recognition","first-page":"12032","article-title":"FIANCEE: Faster inference of adversarial networks via conditional early exits","author":"Polina","year":"2023"},{"key":"10.1016\/j.imavis.2026.105980_b85","series-title":"CAPEEN: Image captioning with early exits and knowledge distillation","author":"Bajpai","year":"2024"},{"key":"10.1016\/j.imavis.2026.105980_b86","doi-asserted-by":"crossref","first-page":"1184","DOI":"10.1109\/LSP.2024.3386110","article-title":"Class based thresholding in early exit semantic segmentation networks","volume":"31","author":"G\u00f6rmez","year":"2024","journal-title":"IEEE Signal Process. Lett."},{"key":"10.1016\/j.imavis.2026.105980_b87","doi-asserted-by":"crossref","first-page":"332","DOI":"10.1109\/TCDS.2023.3274214","article-title":"AdaDet: An adaptive object detection system based on early-exit neural networks","volume":"16","author":"Yang","year":"2024","journal-title":"IEEE Trans. Cogn. Dev. Syst."},{"key":"10.1016\/j.imavis.2026.105980_b88","series-title":"You only look once at anytime (AnytimeYOLO): Analysis and optimization of early-exits for object-detection","author":"Kuhse","year":"2025"},{"key":"10.1016\/j.imavis.2026.105980_b89","series-title":"Adaptive rectangular convolution for remote sensing pansharpening","author":"Wang","year":"2025"},{"key":"10.1016\/j.imavis.2026.105980_b90","doi-asserted-by":"crossref","DOI":"10.1016\/j.patcog.2021.107830","article-title":"Copycat CNN: Are random non-Labeled data enough to steal knowledge from black-box models?","volume":"113","author":"Correia-Silva","year":"2021","journal-title":"Pattern Recognit."},{"key":"10.1016\/j.imavis.2026.105980_b91","doi-asserted-by":"crossref","first-page":"447","DOI":"10.1109\/TWC.2019.2946140","article-title":"Edge AI: On-demand accelerating deep neural network inference via edge computing","volume":"19","author":"Li","year":"2020","journal-title":"IEEE Trans. Wirel. Commun."},{"key":"10.1016\/j.imavis.2026.105980_b92","series-title":"An introduction to convolutional neural networks","author":"O\u2019Shea","year":"2015"},{"key":"10.1016\/j.imavis.2026.105980_b93","doi-asserted-by":"crossref","first-page":"79","DOI":"10.1162\/neco.1991.3.1.79","article-title":"Adaptive mixtures of local experts","volume":"3","author":"Jacobs","year":"1991","journal-title":"Neural Comput."},{"key":"10.1016\/j.imavis.2026.105980_b94","series-title":"Advances in Neural Information Processing Systems","article-title":"ImageNet classification with deep convolutional neural networks","author":"Krizhevsky","year":"2012"},{"key":"10.1016\/j.imavis.2026.105980_b95","series-title":"2015 IEEE conference on computer vision and pattern recognition","first-page":"1","article-title":"Going deeper with convolutions","author":"Szegedy","year":"2015"},{"key":"10.1016\/j.imavis.2026.105980_b96","series-title":"Published as a conference paper at iclr 2018 routing networks: Routing networks: adaptive selection of non-linear functions for multi-task learning","author":"Rosenbaum","year":"2018"},{"key":"10.1016\/j.imavis.2026.105980_b97","article-title":"Dynamic deep neural networks: optimizing accuracy-efficiency trade-offs by selective execution","volume":"32","author":"Liu","year":"2018","journal-title":"Proc. AAAI Conf. Artif. Intell."},{"key":"10.1016\/j.imavis.2026.105980_b98","doi-asserted-by":"crossref","first-page":"730","DOI":"10.1007\/s11263-019-01190-4","article-title":"Convolutional networks with adaptive inference graphs","volume":"128","author":"Veit","year":"2019","journal-title":"Int. J. Comput. Vis."},{"key":"10.1016\/j.imavis.2026.105980_b99","doi-asserted-by":"crossref","first-page":"59","DOI":"10.1007\/s00138-022-01311-z","article-title":"Dynamically throttleable neural networks","volume":"33","author":"Liu","year":"2022","journal-title":"Mach. Vis. Appl."},{"key":"10.1016\/j.imavis.2026.105980_b100","series-title":"Advances in Neural Information Processing Systems","article-title":"The epoch-greedy algorithm for multi-armed bandits with side information","author":"Langford","year":"2007"},{"key":"10.1016\/j.imavis.2026.105980_b101","series-title":"Learning to generate content-aware dynamic detectors","author":"Feng","year":"2020"},{"key":"10.1016\/j.imavis.2026.105980_b102","series-title":"2020 IEEE\/CVF Conference on Computer Vision and Pattern Recognition","first-page":"8550","article-title":"Learning dynamic routing for semantic segmentation","author":"Li","year":"2020"},{"key":"10.1016\/j.imavis.2026.105980_b103","doi-asserted-by":"crossref","first-page":"2400","DOI":"10.1109\/TPAMI.2022.3162528","article-title":"SegBlocks: Block-based dynamic resolution networks for real-time segmentation","volume":"45","author":"Verelst","year":"2023","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"10.1016\/j.imavis.2026.105980_b104","series-title":"Advances in Neural Information Processing Systems","article-title":"Runtime neural pruning","author":"Lin","year":"2017"},{"key":"10.1016\/j.imavis.2026.105980_b105","unstructured":"X. Gao, Y. Zhao, \u0141. Dudziak, R. Mullins, C.z. Xu, Dynamic channel pruning: Feature boosting and suppression, in: International Conference on Learning Representations."},{"key":"10.1016\/j.imavis.2026.105980_b106","series-title":"2019 IEEE\/CVF Conference on Computer Vision and Pattern Recognition","first-page":"9164","article-title":"You look twice: GaterNet for dynamic filter selection in CNNs","author":"Chen","year":"2019"},{"key":"10.1016\/j.imavis.2026.105980_b107","series-title":"Discrete autoencoders for sequence models","author":"Kaiser","year":"2018"},{"key":"10.1016\/j.imavis.2026.105980_b108","doi-asserted-by":"crossref","unstructured":"T. Verelst, T. Tuytelaars, Dynamic convolutions: exploiting spatial sparsity for faster inference, in: Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, 2020, pp. 2320\u20132329.","DOI":"10.1109\/CVPR42600.2020.00239"},{"key":"10.1016\/j.imavis.2026.105980_b109","series-title":"2021 IEEE\/CVF International Conference on Computer Vision","first-page":"5310","article-title":"Dynamic dual gating neural networks","author":"Li","year":"2021"},{"key":"10.1016\/j.imavis.2026.105980_b110","series-title":"2017 IEEE Conference on Computer Vision and Pattern Recognition","first-page":"1790","article-title":"Spatially adaptive computation time for residual networks","author":"Figurnov","year":"2017"},{"key":"10.1016\/j.imavis.2026.105980_b111","first-page":"5700","article-title":"Fractional Skipping: towards finer-grained dynamic CNN inference","volume":"34","author":"Shen","year":"2020","journal-title":"Proc. AAAI Conf. Artif. Intell."},{"key":"10.1016\/j.imavis.2026.105980_b112","series-title":"Proceedings of the European Conference on Computer Vision","first-page":"409","article-title":"SkipNet: Learning dynamic routing in convolutional networks","author":"Wang","year":"2018"},{"key":"10.1016\/j.imavis.2026.105980_b113","doi-asserted-by":"crossref","first-page":"229","DOI":"10.1023\/A:1022672621406","article-title":"Simple statistical gradient-following algorithms for connectionist reinforcement learning","volume":"8","author":"Williams","year":"1992","journal-title":"Mach. Learn."},{"key":"10.1016\/j.imavis.2026.105980_b114","series-title":"EnergyNet: Energy-Efficient dynamic inference","author":"Wang","year":"2018"},{"key":"10.1016\/j.imavis.2026.105980_b115","series-title":"2018 IEEE\/CVF Conference on Computer Vision and Pattern Recognition","first-page":"8817","article-title":"BlockDrop: Dynamic inference paths in residual networks","author":"Wu","year":"2018"},{"key":"10.1016\/j.imavis.2026.105980_b116","series-title":"2022 IEEE\/CVF Conference on Computer Vision and Pattern Recognition","first-page":"12299","article-title":"AdaViT: Adaptive vision transformers for efficient image recognition","author":"Meng","year":"2022"},{"key":"10.1016\/j.imavis.2026.105980_b117","series-title":"Computer Vision \u2013 ECCV 2016","first-page":"516","article-title":"Network of experts for large-scale image categorization","author":"Ahmed","year":"2016"},{"key":"10.1016\/j.imavis.2026.105980_b118","series-title":"Workshop track -ICLR 2017 changing model behavior at test-time using reinforcement learning","author":"Odena","year":"2017"},{"key":"10.1016\/j.imavis.2026.105980_b119","series-title":"2017 IEEE Conference on Computer Vision and Pattern Recognition","first-page":"5085","article-title":"Hard mixtures of experts for large scale weakly supervised vision","author":"Gross","year":"2017"},{"key":"10.1016\/j.imavis.2026.105980_b120","series-title":"2018 IEEE\/CVF Conference on Computer Vision and Pattern Recognition","first-page":"8080","article-title":"HydraNets: Specialized dynamic architectures for efficient inference","author":"Shazeer","year":"2018"},{"key":"10.1016\/j.imavis.2026.105980_b121","series-title":"Proceedings of the 35th Uncertainty in Artificial Intelligence Conference","first-page":"552","article-title":"Deep mixture of experts via shallow embedding","author":"Wang","year":"2020"},{"key":"10.1016\/j.imavis.2026.105980_b122","series-title":"Channel gating neural networks","author":"Hua","year":"2019"},{"key":"10.1016\/j.imavis.2026.105980_b123","doi-asserted-by":"crossref","first-page":"7656","DOI":"10.1109\/TIP.2020.3005508","article-title":"Biased mixtures of experts: enabling computer vision inference under data transfer limitations","volume":"29","author":"Abbas","year":"2020","journal-title":"IEEE Trans. Image Process."},{"key":"10.1016\/j.imavis.2026.105980_b124","doi-asserted-by":"crossref","first-page":"8747","DOI":"10.1109\/TIP.2020.3018269","article-title":"Learning layer-skippable inference network","volume":"29","author":"Jiang","year":"2020","journal-title":"IEEE Trans. Image Process."},{"key":"10.1016\/j.imavis.2026.105980_b125","series-title":"DSelect-k: Differentiable selection in the mixture of experts with applications to multi-task learning","author":"Hazimeh","year":"2021"},{"key":"10.1016\/j.imavis.2026.105980_b126","series-title":"2023 IEEE\/CVF International Conference on Computer Vision","first-page":"90","article-title":"Robust mixture-of-expert training for convolutional neural networks","author":"Zhang","year":"2023"},{"key":"10.1016\/j.imavis.2026.105980_b127","series-title":"Advances in Neural Information Processing Systems","first-page":"8583","article-title":"Scaling vision with sparse mixture of experts","author":"Riquelme","year":"2021"},{"key":"10.1016\/j.imavis.2026.105980_b128","series-title":"Routers in vision mixture of experts: An empirical study","author":"Liu","year":"2024"},{"key":"10.1016\/j.imavis.2026.105980_b129","series-title":"M3ViT: Mixture-of-experts vision transformer for efficient multi-task learning with model-accelerator co-design","author":"Liang","year":"2022"},{"key":"10.1016\/j.imavis.2026.105980_b130","first-page":"58480","article-title":"Mixture of nested experts: Adaptive processing of visual tokens","volume":"37","author":"Jain","year":"2024","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"10.1016\/j.imavis.2026.105980_b131","first-page":"16067","article-title":"MoDE: A mixture-of-experts model with mutual distillation among the experts","volume":"38","author":"Xie","year":"2024","journal-title":"Proc. AAAI Conf. Artif. Intell."},{"key":"10.1016\/j.imavis.2026.105980_b132","series-title":"ViMoE: An empirical study of designing vision mixture-of-experts","author":"Han","year":"2024"},{"key":"10.1016\/j.imavis.2026.105980_b133","series-title":"2020 IEEE\/CVF Conference on Computer Vision and Pattern Recognition Workshops","first-page":"1399","article-title":"Using mixture of expert models to gain insights into semantic segmentation","author":"Pavlitskaya","year":"2020"},{"key":"10.1016\/j.imavis.2026.105980_b134","doi-asserted-by":"crossref","first-page":"623","DOI":"10.1109\/JSTSP.2020.2979669","article-title":"Dual dynamic inference: Enabling more efficient, adaptive, and controllable deep inference","volume":"14","author":"Wang","year":"2020","journal-title":"IEEE J. Sel. Top. Signal Process."},{"key":"10.1016\/j.imavis.2026.105980_b135","series-title":"2021 IFIP\/IEEE 29th International Conference on Very Large Scale Integration","first-page":"1","article-title":"Processor architecture optimization for spatially dynamic neural networks","author":"Colleman","year":"2021"},{"key":"10.1016\/j.imavis.2026.105980_b136","series-title":"2021 IEEE\/CVF Conference on Computer Vision and Pattern Recognition","first-page":"8401","article-title":"Learning dynamic network using a reuse gate function in semi-supervised video object segmentation","author":"Park","year":"2021"},{"key":"10.1016\/j.imavis.2026.105980_b137","series-title":"2021 IEEE\/CVF International Conference on Computer Vision","first-page":"7355","article-title":"Dynamic network quantization for efficient video inference","author":"Sun","year":"2021"},{"key":"10.1016\/j.imavis.2026.105980_b138","series-title":"2021 IEEE\/CVF International Conference on Computer Vision","first-page":"16229","article-title":"Adaptive focus for efficient video recognition","author":"Wang","year":"2021"},{"key":"10.1016\/j.imavis.2026.105980_b139","series-title":"2022 IEEE\/CVF Conference on Computer Vision and Pattern Recognition","first-page":"20030","article-title":"AdaFocus V2: End-to-end training of spatial dynamic networks for video recognition","author":"Wang","year":"2022"},{"key":"10.1016\/j.imavis.2026.105980_b140","series-title":"Computer Vision \u2013 ECCV 2022","first-page":"226","article-title":"AdaFocusV3: On unified spatial-temporal dynamic video recognition","volume":"vol. 13664","author":"Wang","year":"2022"},{"key":"10.1016\/j.imavis.2026.105980_b141","series-title":"Proceedings of the 28th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, vol. 2","first-page":"237","article-title":"DPACS: Hardware accelerated dynamic neural network pruning through algorithm-architecture co-design","author":"Gao","year":"2023"},{"key":"10.1016\/j.imavis.2026.105980_b142","doi-asserted-by":"crossref","first-page":"11374","DOI":"10.1109\/TPAMI.2023.3263619","article-title":"GradMDM: Adversarial attack on dynamic networks","volume":"45","author":"Pan","year":"2023","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"10.1016\/j.imavis.2026.105980_b143","doi-asserted-by":"crossref","DOI":"10.1016\/j.neucom.2025.131216","article-title":"Adaptive layer and token selection for efficient fine-tuning of vision transformers","volume":"654","author":"Devoto","year":"2025","journal-title":"Neurocomputing"},{"key":"10.1016\/j.imavis.2026.105980_b144","series-title":"Computer Vision \u2013 ECCV 2024","first-page":"305","article-title":"DyFADet: Dynamic feature aggregation for temporal action detection","volume":"vol. 15104","author":"Yang","year":"2024"},{"key":"10.1016\/j.imavis.2026.105980_b145","first-page":"3788","article-title":"PSReg: Prior-guided sparse mixture of experts for point cloud registration","volume":"39","author":"Huang","year":"2025","journal-title":"Proc. the AAAI Conf. Artif. Intell."},{"key":"10.1016\/j.imavis.2026.105980_b146","series-title":"Towards accurate and efficient 3D object detection for autonomous driving: A mixture of experts computing system on edge","author":"Liu","year":"2025"},{"key":"10.1016\/j.imavis.2026.105980_b147","doi-asserted-by":"crossref","first-page":"4072","DOI":"10.1109\/TMM.2025.3535339","article-title":"HotMoE: Exploring sparse mixture-of-experts for hyperspectral object tracking","volume":"27","author":"Sun","year":"2025","journal-title":"IEEE Trans. Multimed."},{"key":"10.1016\/j.imavis.2026.105980_b148","series-title":"Multi-modal medical diagnosis via large-small model collaboration","author":"Chen","year":"2025"},{"key":"10.1016\/j.imavis.2026.105980_b149","doi-asserted-by":"crossref","first-page":"56","DOI":"10.1016\/j.patcog.2016.09.007","article-title":"Eye tracking data guided feature selection for image classification","volume":"63","author":"Zhou","year":"2017","journal-title":"Pattern Recognit."},{"key":"10.1016\/j.imavis.2026.105980_b150","series-title":"Advances in Neural Information Processing Systems","first-page":"13937","article-title":"DynamicViT: Efficient vision transformers with dynamic token sparsification","author":"Rao","year":"2021"},{"key":"10.1016\/j.imavis.2026.105980_b151","first-page":"2964","article-title":"Evo-ViT: Slow-fast token evolution for dynamic vision transformer","volume":"36","author":"Xu","year":"2022","journal-title":"Proc. AAAI Conf. Artif. Intell."},{"key":"10.1016\/j.imavis.2026.105980_b152","series-title":"Not all patches are what you need: Expediting vision transformers via token reorganizations","author":"Liang","year":"2022"},{"key":"10.1016\/j.imavis.2026.105980_b153","series-title":"Computer Vision \u2013 ECCV 2022","first-page":"396","article-title":"Adaptive token sampling for efficient vision transformers","author":"Fayyaz","year":"2022"},{"key":"10.1016\/j.imavis.2026.105980_b154","series-title":"SaiT: Sparse Vision transformers through adaptive token pruning","author":"Li","year":"2022"},{"key":"10.1016\/j.imavis.2026.105980_b155","series-title":"2022 IEEE\/CVF Conference on Computer Vision and Pattern Recognition","first-page":"10799","article-title":"A-ViT: Adaptive tokens for efficient vision transformer","author":"Yin","year":"2022"},{"key":"10.1016\/j.imavis.2026.105980_b156","series-title":"2023 IEEE\/CVF Conference on Computer Vision and Pattern Recognition","first-page":"10334","article-title":"Beyond attentive tokens: incorporating token importance and diversity for efficient vision transformers","author":"Long","year":"2023"},{"key":"10.1016\/j.imavis.2026.105980_b157","series-title":"The Eleventh International Conference on Learning Representations","article-title":"Token merging: Your vit but faster","author":"Bolya","year":"2023"},{"key":"10.1016\/j.imavis.2026.105980_b158","series-title":"2023 IEEE\/CVF Conference on Computer Vision and Pattern Recognition","first-page":"2092","article-title":"Joint token pruning and squeezing towards more aggressive compression of vision transformers","author":"Wei","year":"2023"},{"key":"10.1016\/j.imavis.2026.105980_b159","series-title":"2023 IEEE\/CVF International Conference on Computer Vision Workshops","first-page":"838","article-title":"MSViT: Dynamic mixed-scale tokenization for vision transformers","author":"Havtorn","year":"2023"},{"key":"10.1016\/j.imavis.2026.105980_b160","series-title":"IECON 2023- 49th Annual Conference of the IEEE Industrial Electronics Society","first-page":"1","article-title":"Token merging with class importance score","author":"Seol","year":"2023"},{"key":"10.1016\/j.imavis.2026.105980_b161","series-title":"Neighbor patches merging reduces spatial redundancy of nature images","author":"Jiang","year":"2023"},{"key":"10.1016\/j.imavis.2026.105980_b162","series-title":"2024 IEEE\/CVF winter conference on applications of computer vision","first-page":"1372","article-title":"Token fusion: Bridging the gap between token pruning and token merging","author":"Kim","year":"2024"},{"key":"10.1016\/j.imavis.2026.105980_b163","series-title":"2024 IEEE\/CVF Winter Conference on Applications of Computer Vision","first-page":"86","article-title":"GTP-ViT: Efficient vision transformers via graph-based token propagation","author":"Xu","year":"2024"},{"key":"10.1016\/j.imavis.2026.105980_b164","series-title":"AI 2023: Advances in Artificial Intelligence","first-page":"28","article-title":"No token left behind: Efficient vision transformer via dynamic token idling","author":"Xu","year":"2024"},{"key":"10.1016\/j.imavis.2026.105980_b165","doi-asserted-by":"crossref","unstructured":"X. Yuan, H. Fei, J. Baek, Efficient transformer adaptation with soft token merging, in: Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, 2024, pp. 3658\u20133668.","DOI":"10.1109\/CVPRW63382.2024.00369"},{"key":"10.1016\/j.imavis.2026.105980_b166","series-title":"IA-RED2: Interpretability-aware redundancy reduction for vision transformers (supplementary material)","author":"Pan","year":"2021"},{"key":"10.1016\/j.imavis.2026.105980_b167","series-title":"Dynamic transformer networks","author":"Mersha","year":"2022"},{"key":"10.1016\/j.imavis.2026.105980_b168","unstructured":"T. Lei, J. Bai, S. Brahma, J. Ainslie, K. Lee, Y. Zhou, N. Du, V. Zhao, Y. Wu, B. Li, Y. Zhang, M.W. Google, Conditional adapters: Parameter-efficient transfer learning with fast inference, URL: https:\/\/proceedings.neurips.cc\/paper_files\/paper\/2023\/file\/19d7204af519eae9993f7f72377a0ec0-Paper-Conference.pdf."},{"key":"10.1016\/j.imavis.2026.105980_b169","series-title":"Similarity of neural network representations revisited","author":"Kornblith","year":"2019"},{"key":"10.1016\/j.imavis.2026.105980_b170","series-title":"2023 IEEE\/CVF Conference on Computer Vision and Pattern Recognition Workshops","first-page":"4599","article-title":"Token merging for fast stable diffusion","author":"Bolya","year":"2023"},{"key":"10.1016\/j.imavis.2026.105980_b171","series-title":"High-resolution image synthesis with latent diffusion models","author":"Rombach","year":"2022"},{"key":"10.1016\/j.imavis.2026.105980_b172","series-title":"2023 IEEE\/CVF International Conference on Computer Vision","first-page":"10354","article-title":"Efficient video action detection with token dropout and context refinement","author":"Chen","year":"2023"},{"key":"10.1016\/j.imavis.2026.105980_b173","series-title":"2023 IEEE\/CVF Conference on Computer Vision and Pattern Recognition","first-page":"2061","article-title":"SparseViT: Revisiting activation sparsity for efficient high-resolution vision transformer","author":"Chen","year":"2023"},{"key":"10.1016\/j.imavis.2026.105980_b174","series-title":"2023 IEEE\/CVF Conference on Computer Vision and Pattern Recognition","first-page":"23631","article-title":"Content-aware token sharing for efficient semantic segmentation with vision transformers","author":"Lu","year":"2023"},{"key":"10.1016\/j.imavis.2026.105980_b175","series-title":"2024 IEEE\/CVF Winter Conference on Applications of Computer Vision","first-page":"2646","article-title":"Revisiting token pruning for object detection and instance segmentation","author":"Liu","year":"2024"},{"key":"10.1016\/j.imavis.2026.105980_b176","series-title":"2024 IEEE\/CVF Conference on Computer Vision and Pattern Recognition","first-page":"16794","article-title":"Scene adaptive sparse transformer for event-based object detection","author":"Peng","year":"2024"},{"key":"10.1016\/j.imavis.2026.105980_b177","doi-asserted-by":"crossref","first-page":"7043","DOI":"10.1007\/s00521-024-09444-0","article-title":"ATFTrans: Attention-weighted token fusion transformer for robust and efficient object tracking","volume":"36","author":"Xu","year":"2024","journal-title":"Neural Comput. Appl."},{"key":"10.1016\/j.imavis.2026.105980_b178","unstructured":"J. Cheng, X. Yao, X. Yuan, J. Han, Not all tokens matter all the time: Dynamic token aggregation towards efficient detection transformers, in: Forty-Second International Conference on Machine Learning, 2025."},{"key":"10.1016\/j.imavis.2026.105980_b179","doi-asserted-by":"crossref","unstructured":"H. Fang, S. Tang, J. Cao, E. Zhang, F. Tang, T.Y. Lee, Attend to not attended: Structure-then-detail token merging for post-training DiT acceleration, in: Proceedings of the Computer Vision and Pattern Recognition Conference, 2025, pp. 18083\u201318092.","DOI":"10.1109\/CVPR52734.2025.01685"},{"key":"10.1016\/j.imavis.2026.105980_b180","series-title":"Forty-Second International Conference on Machine Learning","article-title":"ToMA: Token merge with attention for diffusion models","author":"Lu","year":"2025"},{"key":"10.1016\/j.imavis.2026.105980_b181","series-title":"2021 IEEE\/CVF International Conference on Computer Vision","first-page":"9992","article-title":"Swin transformer: Hierarchical vision transformer using shifted windows","author":"Liu","year":"2021"},{"key":"10.1016\/j.imavis.2026.105980_b182","series-title":"End-to-End Object Detection with Transformers","first-page":"213","author":"Carion","year":"2020"},{"key":"10.1016\/j.imavis.2026.105980_b183","series-title":"Selective sensor fusion for neural visual-inertial odometry","author":"Chen","year":"2019"},{"key":"10.1016\/j.imavis.2026.105980_b184","series-title":"2016 IEEE\/RSJ International Conference on Intelligent Robots and Systems","first-page":"151","article-title":"Choosing smartly: Adaptive multimodal fusion for object detection in changing environments","author":"Mees","year":"2016"},{"key":"10.1016\/j.imavis.2026.105980_b185","series-title":"2022 ACM\/IEEE 13th International Conference on Cyber-Physical Systems","first-page":"68","article-title":"HydraFusion: Context-aware selective sensor fusion for robust and efficient autonomous vehicle perception","author":"Malawade","year":"2022"},{"key":"10.1016\/j.imavis.2026.105980_b186","series-title":"Proceedings of the 41st IEEE\/ACM International Conference on Computer-Aided Design","first-page":"1","article-title":"Romanus: Robust task offloading in modular multi-sensor autonomous driving systems","author":"Chen","year":"2022"},{"key":"10.1016\/j.imavis.2026.105980_b187","series-title":"2023 IEEE\/CVF International Conference on Computer Vision","first-page":"23498","article-title":"Multi-modal gated mixture of local-to-global experts for dynamic image fusion","author":"Cao","year":"2023"},{"key":"10.1016\/j.imavis.2026.105980_b188","series-title":"2017 IEEE International Conference on Robotics and Automation","first-page":"4644","article-title":"AdapNet: Adaptive semantic segmentation in adverse environmental conditions","author":"Valada","year":"2017"},{"key":"10.1016\/j.imavis.2026.105980_b189","series-title":"2018 IEEE\/RSJ International Conference on Intelligent Robots and Systems","first-page":"3670","article-title":"Modular sensor fusion for semantic segmentation","author":"Blum","year":"2018"},{"key":"10.1016\/j.imavis.2026.105980_b190","first-page":"7414","article-title":"TMFormer: Token merging transformer for brain tumor segmentation with missing modalities","volume":"38","author":"Zhang","year":"2024","journal-title":"Proc. AAAI Conf. Artif. Intell."},{"key":"10.1016\/j.imavis.2026.105980_b191","doi-asserted-by":"crossref","first-page":"571","DOI":"10.1109\/TIV.2018.2874555","article-title":"Estimation of steering angle and collision avoidance for automated driving using deep mixture of experts","volume":"3","author":"John","year":"2018","journal-title":"IEEE Trans. Intell. Veh."},{"key":"10.1016\/j.imavis.2026.105980_b192","doi-asserted-by":"crossref","first-page":"261","DOI":"10.1109\/TCE.2023.3238655","article-title":"MIXO: Mixture Of experts-based visual odometry for multicamera autonomous systems","volume":"69","author":"Morra","year":"2023","journal-title":"IEEE Trans. Consum. Electron."},{"key":"10.1016\/j.imavis.2026.105980_b193","series-title":"CARMA: Context-aware runtime reconfiguration for energy-efficient sensor fusion","author":"Zhang","year":"2023"},{"key":"10.1016\/j.imavis.2026.105980_b194","series-title":"2022 18th International Conference on Distributed Computing in Sensor Systems","first-page":"49","article-title":"SELF-CARE: Selective fusion with context-aware low-power edge computing for stress detection","author":"Rashid","year":"2022"},{"key":"10.1016\/j.imavis.2026.105980_b195","doi-asserted-by":"crossref","first-page":"14114","DOI":"10.1109\/JIOT.2023.3265768","article-title":"Stress detection using context-aware sensor fusion from wearable devices","volume":"10","author":"Rashid","year":"2023","journal-title":"IEEE Internet Things J."},{"key":"10.1016\/j.imavis.2026.105980_b196","doi-asserted-by":"crossref","first-page":"11115","DOI":"10.1109\/TII.2024.3399885","article-title":"Adaptive data fusion for state estimation and control of power grids under attack","volume":"20","author":"Mortlock","year":"2024","journal-title":"IEEE Trans. Ind. Inform."}],"container-title":["Image and Vision Computing"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/api.elsevier.com\/content\/article\/PII:S0262885626000879?httpAccept=text\/xml","content-type":"text\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/api.elsevier.com\/content\/article\/PII:S0262885626000879?httpAccept=text\/plain","content-type":"text\/plain","content-version":"vor","intended-application":"text-mining"}],"deposited":{"date-parts":[[2026,4,27]],"date-time":"2026-04-27T11:19:04Z","timestamp":1777288744000},"score":1,"resource":{"primary":{"URL":"https:\/\/linkinghub.elsevier.com\/retrieve\/pii\/S0262885626000879"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,6]]},"references-count":196,"alternative-id":["S0262885626000879"],"URL":"https:\/\/doi.org\/10.1016\/j.imavis.2026.105980","relation":{},"ISSN":["0262-8856"],"issn-type":[{"value":"0262-8856","type":"print"}],"subject":[],"published":{"date-parts":[[2026,6]]},"assertion":[{"value":"Elsevier","name":"publisher","label":"This article is maintained by"},{"value":"A survey on dynamic neural networks: From computer vision to multi-modal sensor fusion","name":"articletitle","label":"Article Title"},{"value":"Image and Vision Computing","name":"journaltitle","label":"Journal Title"},{"value":"https:\/\/doi.org\/10.1016\/j.imavis.2026.105980","name":"articlelink","label":"CrossRef DOI link to publisher maintained version"},{"value":"article","name":"content_type","label":"Content Type"},{"value":"\u00a9 2026 The Authors. Published by Elsevier B.V.","name":"copyright","label":"Copyright"}],"article-number":"105980"}}