{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,22]],"date-time":"2026-07-22T15:03:04Z","timestamp":1784732584436,"version":"3.55.0"},"reference-count":57,"publisher":"Association for Computing Machinery (ACM)","issue":"8","license":[{"start":{"date-parts":[[2026,7,22]],"date-time":"2026-07-22T00:00:00Z","timestamp":1784678400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/legalcode"}],"funder":[{"DOI":"10.13039\/501100003995","name":"Natural Science Foundation of Anhui Province","doi-asserted-by":"crossref","award":["2508085Y040"],"award-info":[{"award-number":["2508085Y040"]}],"id":[{"id":"10.13039\/501100003995","id-type":"DOI","asserted-by":"crossref"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"crossref","award":["62472141"],"award-info":[{"award-number":["62472141"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"crossref"}]},{"DOI":"10.13039\/501100004739","name":"Youth Innovation Promotion Association CAS","doi-asserted-by":"crossref","id":[{"id":"10.13039\/501100004739","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["ACM Trans. Multimedia Comput. Commun. Appl."],"published-print":{"date-parts":[[2026,8,31]]},"abstract":"<jats:p>Classic object detectors are fundamentally limited by their reliance on a single, static viewpoint, which often suffers from occlusions, challenging scales, and ambiguous perspectives. Video-based object detectors can partially mitigate this problem by aggregating information from multiple views, but they usually work with passively gathered sequences and are in no means guaranteed to provide sufficient information for all objects of interest. On the other hand, active vision methods can seek out better views, yet existing active object detectors typically only try to search for a single optimal perspective and discard valuable information gathered along their path. In this article, we introduce a new paradigm for object detection that unifies active exploration with cumulative spatiotemporal aggregation. We train an embodied agent to intelligently explore its environment, guided by a novel, detection-aware reward function that directly encourages seeking out views that resolve visual ambiguities. To leverage this active exploration, we introduce a robust aggregation pipeline that adeptly harnesses SAM 2\u2019s temporal reasoning capabilities to fuse information from the agent\u2019s entire trajectory into a single, coherent set of detections. Through extensive experiments on AI2-THOR, we demonstrate that our framework provides a consistent and substantial performance uplift when applied to a wide range of state-of-the-art detectors, establishing a strong and versatile new baseline for the next generation of active vision systems.<\/jats:p>","DOI":"10.1145\/3816703","type":"journal-article","created":{"date-parts":[[2026,6,18]],"date-time":"2026-06-18T14:06:29Z","timestamp":1781791589000},"page":"1-20","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Enhancing Object Detection with Active Exploration and Spatiotemporal Aggregation"],"prefix":"10.1145","volume":"22","author":[{"ORCID":"https:\/\/orcid.org\/0009-0002-0365-1602","authenticated-orcid":false,"given":"Peiwei","family":"Li","sequence":"first","affiliation":[{"name":"MoE Key Laboratory of Brain-Inspired Intelligent Perception and Cognition, University of Science and Technology of China, Hefei, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3048-6980","authenticated-orcid":false,"given":"Min","family":"Wang","sequence":"additional","affiliation":[{"name":"IAI-ATCDI Joint R&amp;D Center (Laboratory) on Intelligent Technologies for Comprehensive Transportation, Institute of Artificial Intelligence, Hefei Comprehensive National Science Center, Hefei, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1690-9836","authenticated-orcid":false,"given":"Wengang","family":"Zhou","sequence":"additional","affiliation":[{"name":"EEIS Department, University of Science and Technology of China, Hefei, China and IAI-ATCDI Joint R&amp;D Center (Laboratory) on Intelligent Technologies for Comprehensive Transportation, Institute of Artificial Intelligence, Hefei Comprehensive National Science Center, Hefei, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9643-4601","authenticated-orcid":false,"given":"Yufei","family":"Yin","sequence":"additional","affiliation":[{"name":"Hangzhou Dianzi University, Hangzhou, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0007-9652-7972","authenticated-orcid":false,"given":"Yebo","family":"Bao","sequence":"additional","affiliation":[{"name":"Anhui Transport Consulting &amp; Design Institute Co., Ltd., Hefei, China and IAI-ATCDI Joint R&amp;D Center (Laboratory) on Intelligent Technologies for Comprehensive Transportation, Institute of Artificial Intelligence, Hefei Comprehensive National Science Center, Hefei, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0002-8498-5056","authenticated-orcid":false,"given":"Guodong","family":"Shen","sequence":"additional","affiliation":[{"name":"Anhui Transport Consulting &amp; Design Institute Co., Ltd., Hefei, China and IAI-ATCDI Joint R&amp;D Center (Laboratory) on Intelligent Technologies for Comprehensive Transportation, Institute of Artificial Intelligence, Hefei Comprehensive National Science Center, Hefei, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2188-3028","authenticated-orcid":false,"given":"Houqiang","family":"Li","sequence":"additional","affiliation":[{"name":"EEIS Department, University of Science and Technology of China, Hefei, China and IAI-ATCDI Joint R&amp;D Center (Laboratory) on Intelligent Technologies for Comprehensive Transportation, Institute of Artificial Intelligence, Hefei Comprehensive National Science Center, Hefei, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"320","published-online":{"date-parts":[[2026,7,22]]},"reference":[{"key":"e_1_3_2_2_2","doi-asserted-by":"publisher","DOI":"10.1007\/BF00133571"},{"key":"e_1_3_2_3_2","doi-asserted-by":"publisher","DOI":"10.1109\/ICIP.2016.7533003"},{"key":"e_1_3_2_4_2","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58452-8_13"},{"key":"e_1_3_2_5_2","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01599"},{"key":"e_1_3_2_6_2","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2005.177"},{"key":"e_1_3_2_7_2","first-page":"7023","volume-title":"Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition","author":"Deng Jiajun","year":"2019","unstructured":"Jiajun Deng, Yingwei Pan, Ting Yao, Wengang Zhou, Houqiang Li, and Tao Mei. 2019. Relation distillation networks for video object detection. In Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, 7023\u20137032."},{"key":"e_1_3_2_8_2","doi-asserted-by":"publisher","DOI":"10.1109\/34.982896"},{"key":"e_1_3_2_9_2","doi-asserted-by":"publisher","unstructured":"Shuangrui Ding Rui Qian Xiaoyi Dong Pan Zhang Yuhang Zang Yuhang Cao Yuwei Guo Dahua Lin and Jiaqi Wang. 2024. SAM2Long: Enhancing SAM 2 for long video segmentation with a training-free memory tree. In 2025 IEEE\/CVF International Conference on Computer Vision (ICCV). IEEE. Retrieved from 10.1109\/iccv51701.2025.01264","DOI":"10.1109\/iccv51701.2025.01264"},{"key":"e_1_3_2_10_2","doi-asserted-by":"crossref","unstructured":"Wenhao Ding Nathalie Majcherczyk Mohit Deshpande Xuewei Qi Ding Zhao Rajasimman Madhivanan and Arnie Sen. 2023. Learning to view: Decision transformers for active object detection. arXiv:2301.09544. Retrieved from https:\/\/arxiv.org\/abs\/2301.09544","DOI":"10.1109\/ICRA48891.2023.10160946"},{"key":"e_1_3_2_11_2","first-page":"1","volume-title":"Proceedings of the British Machine Vision Conference","author":"Dunn Enrique","year":"2009","unstructured":"Enrique Dunn and Jan-Michael Frahm. 2009. Next best view planning for active model improvement. In Proceedings of the British Machine Vision Conference, 1\u201311."},{"key":"e_1_3_2_12_2","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-009-0275-4"},{"key":"e_1_3_2_13_2","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01551"},{"key":"e_1_3_2_14_2","doi-asserted-by":"crossref","unstructured":"Zhaoyuan Fang Ayush Jain Gabriel Sarch Adam W. Harley and Katerina Fragkiadaki. 2021. Move to see better: Self-improving embodied object detection. arXiv:2012.00057. Retrieved from https:\/\/arxiv.org\/abs\/2012.00057","DOI":"10.5244\/C.35.174"},{"key":"e_1_3_2_15_2","first-page":"1126","volume-title":"Proceedings of the International Conference on Machine Learning","author":"Finn Chelsea","year":"2017","unstructured":"Chelsea Finn, Pieter Abbeel, and Sergey Levine. 2017. Model-agnostic meta-learning for fast adaptation of deep networks. In Proceedings of the International Conference on Machine Learning. PMLR, 1126\u20131135."},{"key":"e_1_3_2_16_2","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00550"},{"key":"e_1_3_2_17_2","unstructured":"Wei Han Pooya Khorrami Tom Le Paine Prajit Ramachandran Mohammad Babaeizadeh Honghui Shi Jianan Li Shuicheng Yan and Thomas S. Huang. 2016. Seq-NMS for video object detection. arXiv:1602.08465. Retrieved from https:\/\/arxiv.org\/abs\/1602.08465"},{"key":"e_1_3_2_18_2","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"e_1_3_2_19_2","doi-asserted-by":"publisher","DOI":"10.1145\/3474085.3475285"},{"key":"e_1_3_2_20_2","unstructured":"Yicong Hong Qi Wu Yuankai Qi Cristian Rodriguez-Opazo and Stephen Gould. 2020. A recurrent vision-and-language BERT for navigation. arXiv:2011.13922. Retrieved from https:\/\/arxiv.org\/abs\/2011.13922"},{"key":"e_1_3_2_21_2","doi-asserted-by":"publisher","DOI":"10.1145\/3640465"},{"key":"e_1_3_2_22_2","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-46454-1_30"},{"key":"e_1_3_2_23_2","unstructured":"Junjie Jiang Zelin Wang Manqi Zhao Yin Li and DongSheng Jiang. 2025. SAM2MOT: A novel paradigm of multi-object tracking by segmentation. arXiv:2504.04519. Retrieved from https:\/\/arxiv.org\/abs\/2504.04519"},{"key":"e_1_3_2_24_2","unstructured":"Glenn Jocher Jing Qiu and Ayush Chaurasia. 2023. Ultralytics YOLO. Retrieved from https:\/\/github.com\/ultralytics\/ultralytics"},{"key":"e_1_3_2_25_2","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.00371"},{"key":"e_1_3_2_26_2","unstructured":"Eric Kolve Roozbeh Mottaghi Winson Han Eli VanderBilt Luca Weihs Alvaro Herrasti Matt Deitke Kiana Ehsani Daniel Gordon Yuke Zhu et al. 2022. AI2-THOR: An interactive 3D environment for visual AI. arXiv:1712.05474. Retrieved from https:\/\/arxiv.org\/abs\/1712.05474"},{"key":"e_1_3_2_27_2","first-page":"14840","volume-title":"Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition","author":"Kotar Klemen","year":"2022","unstructured":"Klemen Kotar and Roozbeh Mottaghi. 2022. Interactron: Embodied adaptive object detection. In Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition. IEEE, 14840\u201314849."},{"key":"e_1_3_2_28_2","first-page":"5031","volume-title":"Proceedings of the International Conference on Robotics and Automation","author":"Krainin Michael","year":"2011","unstructured":"Michael Krainin, Brian Curless, and Dieter Fox. 2011. Autonomous generation of complete 3D object models using next best view manipulation planning. In Proceedings of the International Conference on Robotics and Automation. IEEE, 5031\u20135037."},{"key":"e_1_3_2_29_2","doi-asserted-by":"publisher","DOI":"10.1145\/3706574"},{"key":"e_1_3_2_30_2","doi-asserted-by":"publisher","DOI":"10.1109\/LRA.2025.3580331"},{"key":"e_1_3_2_31_2","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-10602-1_48"},{"key":"e_1_3_2_32_2","first-page":"5686","volume-title":"Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition","author":"Liu Mason","year":"2018","unstructured":"Mason Liu and Menglong Zhu. 2018. Mobile video object detection with temporally-aware feature mAPs. In Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, 5686\u20135695."},{"key":"e_1_3_2_33_2","volume-title":"Proceedings of the International Conference on Learning Representations","author":"Liu Shilong","year":"2022","unstructured":"Shilong Liu, Feng Li, Hao Zhang, Xiao Yang, Xianbiao Qi, Hang Su, Jun Zhu, and Lei Zhang. 2022. DAB-DETR: Dynamic anchor boxes are better queries for DETR. In Proceedings of the International Conference on Learning Representations."},{"key":"e_1_3_2_34_2","unstructured":"Ilya Loshchilov and Frank Hutter. 2017. Decoupled weight decay regularization. arXiv:1711.05101. Retrieved from https:\/\/arxiv.org\/abs\/1711.05101"},{"key":"e_1_3_2_35_2","unstructured":"Feiyu Pan Hao Fang Runmin Cong Wei Zhang and Xiankai Lu. 2024. Video object segmentation via SAM 2: The 4th solution for LSVOS challenge VOS track. arXiv:2408.10125. Retrieved from https:\/\/arxiv.org\/abs\/2408.10125"},{"key":"e_1_3_2_36_2","doi-asserted-by":"publisher","DOI":"10.1145\/3637214"},{"key":"e_1_3_2_37_2","unstructured":"Nikhila Ravi Valentin Gabeur Yuan-Ting Hu Ronghang Hu Chaitanya Ryali Tengyu Ma Haitham Khedr Roman R\u00e4dle Chloe Rolland Laura Gustafson et al. 2024. SAM 2: Segment anything in images and videos. arXiv:2408.00714. Retrieved from https:\/\/arxiv.org\/abs\/2408.00714"},{"key":"e_1_3_2_38_2","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.91"},{"key":"e_1_3_2_39_2","first-page":"91","volume-title":"Advances in Neural Information Processing Systems","volume":"28","author":"Ren Shaoqing","year":"2015","unstructured":"Shaoqing Ren, Kaiming He, Ross Girshick, and Jian Sun. 2015. Faster r-CNN: Towards real-time object detection with region proposal networks. In Advances in Neural Information Processing Systems, Vol. 28, 91\u201399."},{"key":"e_1_3_2_40_2","first-page":"8355","volume-title":"Proceedings of the Winter Conference on Applications of Computer Vision","author":"Ren Simiao","year":"2024","unstructured":"Simiao Ren, Francesco Luzi, Saad Lahrichi, Kaleb Kassaw, Leslie M. Collins, Kyle Bradbury, and Jordan M. Malof. 2024. Segment anything, from space? In Proceedings of the Winter Conference on Applications of Computer Vision, 8355\u20138365."},{"key":"e_1_3_2_41_2","unstructured":"John Schulman Filip Wolski Prafulla Dhariwal Alec Radford and Oleg Klimov. 2017. Proximal policy optimization algorithms. arXiv:1707.06347. Retrieved from https:\/\/arxiv.org\/abs\/1707.06347"},{"key":"e_1_3_2_42_2","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v38i5.28281"},{"key":"e_1_3_2_43_2","first-page":"2254","volume-title":"Proceedings of the AAAI Conference on Artificial Intelligence","volume":"37","author":"Shi Yuheng","year":"2023","unstructured":"Yuheng Shi, Naiyan Wang, and Xiaojie Guo. 2023. Yolov: Making still image object detectors great at video object detection. In Proceedings of the AAAI Conference on Artificial Intelligence, Vol. 37, 2254\u20132262."},{"key":"e_1_3_2_44_2","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.00252"},{"key":"e_1_3_2_45_2","volume-title":"Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition","volume":"1","author":"Viola Paul","year":"2001","unstructured":"Paul Viola and Michael Jones. 2001. Rapid object detection using a boosted Cascade of simple features. In Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, Vol. 1. IEEE."},{"key":"e_1_3_2_46_2","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-20074-8_42"},{"key":"e_1_3_2_47_2","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-01270-0_3"},{"key":"e_1_3_2_48_2","doi-asserted-by":"publisher","DOI":"10.1145\/3584362"},{"key":"e_1_3_2_49_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.neucom.2021.09.037"},{"key":"e_1_3_2_50_2","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00213"},{"key":"e_1_3_2_51_2","unstructured":"Yunhan Yang Xiaoyang Wu Tong He Hengshuang Zhao and Xihui Liu. 2023. SAM3D: Segment anything in 3D scenes. arXiv:2306.03908. Retrieved from https:\/\/arxiv.org\/abs\/2306.03908"},{"key":"e_1_3_2_52_2","unstructured":"Tao Yu Runseng Feng Ruoyu Feng Jinming Liu Xin Jin Wenjun Zeng and Zhibo Chen. 2023. Inpaint anything: Segment anything meets image inpainting. arXiv:2304.06790. Retrieved from https:\/\/arxiv.org\/abs\/2304.06790"},{"key":"e_1_3_2_53_2","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-20047-2_1"},{"key":"e_1_3_2_54_2","unstructured":"Xu Zhao Wenchao Ding Yongqi An Yinglong Du Tao Yu Min Li Ming Tang and Jinqiao Wang. 2023. Fast segment anything. arXiv:2306.12156. Retrieved from https:\/\/arxiv.org\/abs\/2306.12156"},{"key":"e_1_3_2_55_2","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2022.3223955"},{"key":"e_1_3_2_56_2","first-page":"7210","volume-title":"Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition","author":"Zhu Xizhou","year":"2018","unstructured":"Xizhou Zhu, Jifeng Dai, Lu Yuan, and Yichen Wei. 2018. Towards high performance video object detection. In Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, 7210\u20137218."},{"key":"e_1_3_2_57_2","unstructured":"Xizhou Zhu Weijie Su Lewei Lu Bin Li Xiaogang Wang and Jifeng Dai. 2021. Deformable DETR: Deformable transformers for end-to-end object detection. arXiv:2010.04159. Retrieved from https:\/\/arxiv.org\/abs\/2010.04159"},{"key":"e_1_3_2_58_2","first-page":"408","volume-title":"Proceedings of the IEEE\/CVF International Conference on Computer Vision","author":"Zhu Xizhou","year":"2017","unstructured":"Xizhou Zhu, Yujie Wang, Jifeng Dai, Lu Yuan, and Yichen Wei. 2017. Flow-guided feature aggregation for video object detection. In Proceedings of the IEEE\/CVF International Conference on Computer Vision, 408\u2013417."}],"container-title":["ACM Transactions on Multimedia Computing, Communications, and Applications"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3816703","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,22]],"date-time":"2026-07-22T14:27:45Z","timestamp":1784730465000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3816703"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,22]]},"references-count":57,"journal-issue":{"issue":"8","published-print":{"date-parts":[[2026,8,31]]}},"alternative-id":["10.1145\/3816703"],"URL":"https:\/\/doi.org\/10.1145\/3816703","relation":{},"ISSN":["1551-6857","1551-6865"],"issn-type":[{"value":"1551-6857","type":"print"},{"value":"1551-6865","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,22]]},"assertion":[{"value":"2025-10-17","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-04-29","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-07-22","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}