{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,30]],"date-time":"2026-07-30T13:18:13Z","timestamp":1785417493913,"version":"3.56.0"},"reference-count":50,"publisher":"MDPI AG","issue":"7","license":[{"start":{"date-parts":[[2026,7,18]],"date-time":"2026-07-18T00:00:00Z","timestamp":1784332800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100007276","name":"ASELSAN Inc.","doi-asserted-by":"crossref","id":[{"id":"10.13039\/501100007276","id-type":"DOI","asserted-by":"crossref"}]},{"DOI":"10.13039\/501100004410","name":"The Scientific and Technological Research Council of Turkiye","doi-asserted-by":"publisher","award":["5239903"],"award-info":[{"award-number":["5239903"]}],"id":[{"id":"10.13039\/501100004410","id-type":"DOI","asserted-by":"publisher"}]},{"award":["5239903"],"award-info":[{"award-number":["5239903"]}],"id":[{"id":"https:\/\/ror.org\/04w9kkr77","id-type":"ROR","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100007504","name":"Istanbul Technical University","doi-asserted-by":"publisher","award":["ITU-BAP MGA-2024-45372"],"award-info":[{"award-number":["ITU-BAP MGA-2024-45372"]}],"id":[{"id":"10.13039\/501100007504","id-type":"DOI","asserted-by":"publisher"}]},{"award":["ITU-BAP MGA-2024-45372"],"award-info":[{"award-number":["ITU-BAP MGA-2024-45372"]}],"id":[{"id":"https:\/\/ror.org\/059636586","id-type":"ROR","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["J. Imaging"],"abstract":"<jats:p>The proliferation of high-definition video data necessitates highly efficient processing pipelines for real-time edge analytics. However, traditional object detection architectures rely exclusively on pixel-domain inputs, which renders the computationally prohibitive decoding phase a latency bottleneck. In this paper, we propose a novel dual-phase framework designed to achieve fast and efficient object detection directly within the partially decoded compressed-domain data. First, we introduce a partial decoding paradigm featuring the Low-Frequency Spectral Prioritization method on the encoder side. By systematically discarding high-frequency residual coefficients and retaining only a sparse subset of fundamental spatial frequencies, this method dramatically reduces transmission payloads and accelerates the standard decoding process. Second, to recover the structural fidelity lost due to the intentional omission of residual data, we employ a multi-granularity cross-domain knowledge distillation architecture. This strategy aligns global contextual features, foreground boundary attention maps, and final response logits, transferring rich representational capacities from a high-performing pixel-domain teacher network to a lightweight compressed-domain student network. Comprehensive experiments utilizing RetinaNet, FCOS, and GFL object detection networks on the COCO-mini dataset demonstrate the superiority of the proposed framework. By retaining fundamental residual coefficients within the HEVC pipeline, the proposed method reduces average decoding latency while improving the mAP score by +0.96% over the conventional fully decoded pixel-domain baseline on the COCO-mini dataset.<\/jats:p>","DOI":"10.3390\/jimaging12070325","type":"journal-article","created":{"date-parts":[[2026,7,20]],"date-time":"2026-07-20T06:09:09Z","timestamp":1784527749000},"page":"325","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Efficient Object Detection in Compressed Domain by Exploiting Knowledge Distillation from Pixel Domain"],"prefix":"10.3390","volume":"12","author":[{"ORCID":"https:\/\/orcid.org\/0009-0006-0401-9114","authenticated-orcid":false,"given":"Serhat","family":"Dikyar","sequence":"first","affiliation":[{"name":"ASELSAN Inc., Ankara 06370, T\u00fcrkiye"},{"name":"Signal Processing for Computational Intelligence Research Group (SP4CING), \u0130stanbul Technical University, \u0130stanbul 34469, T\u00fcrkiye"},{"name":"Department of Electronics and Communication Engineering, \u0130stanbul Technical University, \u0130stanbul 34469, T\u00fcrkiye"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4406-2783","authenticated-orcid":false,"given":"Behcet Ugur","family":"Toreyin","sequence":"additional","affiliation":[{"name":"Signal Processing for Computational Intelligence Research Group (SP4CING), \u0130stanbul Technical University, \u0130stanbul 34469, T\u00fcrkiye"},{"name":"Department of Artificial Intelligence and Data Engineering, \u0130stanbul Technical University, \u0130stanbul 34469, T\u00fcrkiye"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,7,18]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"560","DOI":"10.1109\/TCSVT.2003.815165","article-title":"Overview of the H.264\/AVC video coding standard","volume":"13","author":"Wiegand","year":"2003","journal-title":"IEEE Trans. Circuits Syst. Video Technol."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"1649","DOI":"10.1109\/TCSVT.2012.2221191","article-title":"Overview of the High Efficiency Video Coding (HEVC) standard","volume":"22","author":"Sullivan","year":"2012","journal-title":"IEEE Trans. Circuits Syst. Video Technol."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"3736","DOI":"10.1109\/TCSVT.2021.3101953","article-title":"Overview of the Versatile Video Coding (VVC) Standard and its Applications","volume":"31","author":"Bross","year":"2021","journal-title":"IEEE Trans. Circuits Syst. Video Technol."},{"key":"ref_4","doi-asserted-by":"crossref","unstructured":"Mart\u00ednez-Rach, M.O., Migall\u00f3n, H., L\u00f3pez-Granado, O., Galiano, V., and Malumbres, M.P. (2021). Performance Overview of the Latest Video Coding Proposals: HEVC, JEM and VVC. J. Imaging, 7.","DOI":"10.3390\/jimaging7020039"},{"key":"ref_5","doi-asserted-by":"crossref","unstructured":"Biswas, S.D., Soufleri, E., Roy, A., and Roy, K. (2026, January 6\u201310). Learning Unified Spatio-temporal Representations for Efficient Compressed Video Understanding. Proceedings of the 2026 IEEE\/CVF Winter Conference on Applications of Computer Vision (WACV), Tucson, Arizona.","DOI":"10.1109\/WACV61042.2026.00462"},{"key":"ref_6","doi-asserted-by":"crossref","unstructured":"Zhao, L., Zhao, D., Fan, X., and He, Z. (2016). HEVC compressed domain moving object detection and classification. Proceedings of the IEEE International Symposium on Circuits and Systems (ISCAS), IEEE.","DOI":"10.1109\/ISCAS.2016.7538966"},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"95087","DOI":"10.1109\/ACCESS.2021.3092938","article-title":"Vehicle license plate detector in compressed domain","volume":"9","year":"2021","journal-title":"IEEE Access"},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"140","DOI":"10.1016\/j.jvcir.2015.12.011","article-title":"Compressed domain video saliency detection using global and local spatiotemporal features","volume":"35","author":"Lee","year":"2016","journal-title":"J. Vis. Commun. Image Represent."},{"key":"ref_9","doi-asserted-by":"crossref","unstructured":"Fang, X., Liu, D., Zhou, P., and Nan, G. (2023, January 17\u201324). You Can Ground Earlier than See: An Effective and Efficient Pipeline for Temporal Sentence Grounding in Compressed Videos. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Vancouver, BC, Canada.","DOI":"10.1109\/CVPR52729.2023.00242"},{"key":"ref_10","doi-asserted-by":"crossref","unstructured":"Kim, N., Ha, S.J., and Kang, J.W. (2021, January 10\u201317). Video Question Answering Using Language-Guided Deep Compressed-Domain Video Feature. Proceedings of the IEEE\/CVF International Conference on Computer Vision (ICCV), Montreal, QC, Canada.","DOI":"10.1109\/ICCV48922.2021.00173"},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Nguyen, D.V., and Choi, J. (2020). Toward scalable video analytics using compressed-domain features at the edge. Appl. Sci., 10.","DOI":"10.3390\/app10186391"},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"Chen, L., Sun, H., Katto, J., Zeng, X., and Fan, Y. (2021). Fast object detection in HEVC intra compressed domain. Proceedings of the 29th European Signal Processing Conference (EUSIPCO), IEEE.","DOI":"10.23919\/EUSIPCO54536.2021.9616315"},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"2836","DOI":"10.1109\/TDSC.2023.3318975","article-title":"Secure deep learning framework for moving object detection in compressed video","volume":"21","author":"Tian","year":"2024","journal-title":"IEEE Trans. Dependable Secur. Comput."},{"key":"ref_14","unstructured":"Hinton, G., Vinyals, O., and Dean, J. (2015). Distilling the knowledge in a neural network. arXiv."},{"key":"ref_15","doi-asserted-by":"crossref","unstructured":"Shehzadi, T., Noor, R., Ifza, I., Liwicki, M., Stricker, D., and Afzal, M.Z. (2026). Knowledge Distillation in Object Detection: A Survey from CNN to Transformer. Sensors, 26.","DOI":"10.3390\/s26010292"},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Gupta, S., Hoffman, J., and Malik, J. (2016, January 27\u201330). Cross Modal Distillation for Supervision Transfer. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, USA.","DOI":"10.1109\/CVPR.2016.309"},{"key":"ref_17","doi-asserted-by":"crossref","unstructured":"Liu, Y., Cao, J., Bai, W., Li, B., and Hu, W. (2023). Learning from the Raw Domain: Cross Modality Distillation for Compressed Video Action Recognition. Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), IEEE.","DOI":"10.1109\/ICASSP49357.2023.10097076"},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Lu, H., and Deng, Z. (2021). A boundary-aware distillation network for compressed video semantic segmentation. Proceedings of the 25th International Conference on Pattern Recognition (ICPR), IEEE.","DOI":"10.1109\/ICPR48806.2021.9412821"},{"key":"ref_19","first-page":"64","article-title":"Transformers in Small Object Detection: A Benchmark and Survey of State-of-the-Art","volume":"58","author":"Rashidi","year":"2025","journal-title":"ACM Comput. Surv."},{"key":"ref_20","doi-asserted-by":"crossref","unstructured":"Mi, Q., Chao, J., Chen, A., Zhang, K., and Lai, J. (2026). YOLO11s-UAV: An Advanced Algorithm for Small Object Detection in UAV Aerial Imagery. J. Imaging, 12.","DOI":"10.3390\/jimaging12020069"},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Wang, J., Chen, K., Xu, R., Liu, Z., Loy, C.C., and Lin, D. (2019). CARAFE: Content-Aware ReAssembly of FEatures. Proceedings of the IEEE\/CVF International Conference on Computer Vision (ICCV), IEEE.","DOI":"10.1109\/ICCV.2019.00310"},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"116675","DOI":"10.1016\/j.image.2022.116675","article-title":"Focus-and-Detect: A small object detection framework for aerial images","volume":"104","author":"Koyun","year":"2022","journal-title":"Signal Process. Image Commun."},{"key":"ref_23","doi-asserted-by":"crossref","unstructured":"Yang, F., Fan, H., Chu, P., Blasch, E., and Ling, H. (2019). Clustered object detection in aerial images. Proceedings of the IEEE\/CVF International Conference on Computer Vision (ICCV), IEEE.","DOI":"10.1109\/ICCV.2019.00840"},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"2754","DOI":"10.1109\/TETCI.2024.3442819","article-title":"Low-Light Salient Object Detection Meets the Small Size","volume":"9","author":"Wang","year":"2024","journal-title":"IEEE Trans. Emerg. Top. Comput. Intell."},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"109938","DOI":"10.1016\/j.knosys.2022.109938","article-title":"Salient object detection in low-light images via functional optimization-inspired feature polishing","volume":"257","author":"Yue","year":"2022","journal-title":"Knowl.-Based Syst."},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"7285","DOI":"10.1007\/s00371-025-03805-z","article-title":"Context-aware target texture perturbation attack for concealed object detection","volume":"41","author":"Zhang","year":"2025","journal-title":"Vis. Comput."},{"key":"ref_27","doi-asserted-by":"crossref","unstructured":"Gao, R., Guo, Q., Juefei-Xu, F., Yu, H., Fu, H., Feng, W., Liu, Y., and Wang, S. (2022). Can You Spot the Chameleon? Adversarially Camouflaging Images from Co-Salient Object Detection. Proceedings of the 2022 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR), IEEE.","DOI":"10.1109\/CVPR52688.2022.00219"},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"e7","DOI":"10.1017\/ATSIP.2016.18","article-title":"Moving object detection in the H.264\/AVC compressed domain","volume":"5","author":"Laumer","year":"2016","journal-title":"APSIPA Trans. Signal Inf. Process."},{"key":"ref_29","doi-asserted-by":"crossref","unstructured":"Jaballah, S., and Larabi, M.-C. (2019). Fast object detection in H.264\/AVC and HEVC compressed domains for video surveillance. Proceedings of the 8th European Workshop on Visual Information Processing (EUVIP), IEEE.","DOI":"10.1109\/EUVIP47703.2019.8946214"},{"key":"ref_30","doi-asserted-by":"crossref","unstructured":"Berato\u011flu, M.S., and T\u00f6reyin, B.U. (2026). Residual-Free Image Reconstruction from HEVC for Compressed-Domain Visual Analytics. Proceedings of the IEEE International Symposium on Circuits and Systems (ISCAS), IEEE.","DOI":"10.1109\/ISCAS66217.2026.11562979"},{"key":"ref_31","doi-asserted-by":"crossref","unstructured":"Dror, I., and Hadar, O. (2025). Improved Perceptual Quality of Traffic Signs and Lights for the Teleoperation of Autonomous Vehicle Remote Driving via Multi-Category Region of Interest Video Compression. Entropy, 27.","DOI":"10.3390\/e27070674"},{"key":"ref_32","doi-asserted-by":"crossref","unstructured":"Zain, A., Fan, Y., and Zhou, J. (2026). On Demand Secure Scalable Video Streaming for Both Human and Machine Applications. Sensors, 26.","DOI":"10.3390\/s26041285"},{"key":"ref_33","doi-asserted-by":"crossref","unstructured":"Wang, S., Lu, H., and Deng, Z. (2019). Fast object detection in compressed video. Proceedings of the IEEE\/CVF International Conference on Computer Vision (ICCV), IEEE.","DOI":"10.1109\/ICCV.2019.00720"},{"key":"ref_34","doi-asserted-by":"crossref","first-page":"3569","DOI":"10.1109\/TCSVT.2020.2971641","article-title":"Real time video object segmentation in compressed domain","volume":"31","author":"Tan","year":"2021","journal-title":"IEEE Trans. Circuits Syst. Video Technol."},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"7724","DOI":"10.1109\/TPAMI.2020.3024646","article-title":"TapLab: A fast framework for semantic video segmentation tapping into compressed-domain knowledge","volume":"44","author":"Feng","year":"2022","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"ref_36","doi-asserted-by":"crossref","unstructured":"Zhang, B., Wang, L., Wang, Z., Qiao, Y., and Wang, H. (2016). Real-time action recognition with enhanced motion vector CNNs. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), IEEE.","DOI":"10.1109\/CVPR.2016.297"},{"key":"ref_37","unstructured":"Huang, S., Lin, X., Karaman, S., and Chang, S.F. (2019). Flow-distilled IP two-stream networks for compressed video action recognition. arXiv."},{"key":"ref_38","doi-asserted-by":"crossref","unstructured":"Huo, Y., Xu, X., Lu, Y., Niu, Y., Ding, M., Lu, Z., Xiang, T., and Wen, J.R. (2020). Lightweight action recognition in compressed videos. Proceedings of the European Conference on Computer Vision (ECCV) Workshops, Springer International Publishing.","DOI":"10.1007\/978-3-030-66096-3_24"},{"key":"ref_39","doi-asserted-by":"crossref","first-page":"7926","DOI":"10.1109\/TIP.2021.3112008","article-title":"Joint feature optimization and fusion for compressed action recognition","volume":"30","author":"Li","year":"2021","journal-title":"IEEE Trans. Image Process."},{"key":"ref_40","doi-asserted-by":"crossref","first-page":"4108","DOI":"10.1109\/TCSVT.2023.3240391","article-title":"Unified architecture adaptation for compressed domain semantic inference","volume":"33","author":"Duan","year":"2023","journal-title":"IEEE Trans. Circuits Syst. Video Technol."},{"key":"ref_41","doi-asserted-by":"crossref","unstructured":"Liu, J., Sun, H., and Katto, J. (2022). Improving multiple machine vision tasks in the compressed domain. Proceedings of the 26th International Conference on Pattern Recognition (ICPR), IEEE.","DOI":"10.1109\/ICPR56361.2022.9956532"},{"key":"ref_42","doi-asserted-by":"crossref","unstructured":"Keser, R.K., Berato\u011flu, M.S., and T\u00f6reyin, B.U. (2024). Generated compressed domain images to the rescue: Cross distillation from compressed domain to pixel domain. Proceedings of the IEEE International Symposium on Circuits and Systems (ISCAS), IEEE.","DOI":"10.1109\/ISCAS58744.2024.10558450"},{"key":"ref_43","unstructured":"Li, L., Bao, Y., Dong, P., Yang, C., Li, A., Luo, W., Liu, Q., Xue, W., and Guo, Y. (2024, January 21\u201327). DetKDS: Knowledge distillation search for object detectors. Proceedings of the 41st International Conference on Machine Learning (ICML), Vienna, Austria."},{"key":"ref_44","doi-asserted-by":"crossref","unstructured":"Lin, T.-Y., Maire, M., Belongie, S., Hays, J., Perona, P., Ramanan, D., Doll\u00e1r, P., and Zitnick, C.L. (2014). Microsoft COCO: Common objects in context. Computer Vision\u2014ECCV 2014, Springer.","DOI":"10.1007\/978-3-319-10602-1_48"},{"key":"ref_45","doi-asserted-by":"crossref","unstructured":"Samet, N., Hicsonmez, S., and Akbas, E. (2020). HoughNet: Integrating near and long-range evidence for bottom-up object detection. Proceedings of the European Conference on Computer Vision (ECCV), Springer International Publishing.","DOI":"10.1007\/978-3-030-58595-2_25"},{"key":"ref_46","unstructured":"JCT-VC (2026, July 12). HEVC Reference Software, Version HM 18.0. Available online: https:\/\/vcgit.hhi.fraunhofer.de\/jvet\/HM\/-\/releases\/HM-18.0."},{"key":"ref_47","doi-asserted-by":"crossref","unstructured":"Lin, T.-Y., Goyal, P., Girshick, R., He, K., and Doll\u00e1r, P. (2017). Focal loss for dense object detection. Proceedings of the IEEE International Conference on Computer Vision (ICCV), IEEE.","DOI":"10.1109\/ICCV.2017.324"},{"key":"ref_48","unstructured":"Chen, K., Wang, J., Pang, J., Cao, Y., Xiong, Y., Li, X., Sun, S., Feng, W., Liu, Z., and Xu, J. (2019). MMDetection: Open MMLab Detection Toolbox and Benchmark. arXiv."},{"key":"ref_49","unstructured":"Tian, Z., Shen, C., Chen, H., and He, T. (November, January 27). FCOS: Fully convolutional one-stage object detection. Proceedings of the IEEE\/CVF International Conference on Computer Vision (ICCV), Seoul, Republic of Korea."},{"key":"ref_50","unstructured":"Li, X., Wang, W., Wu, L., Chen, S., Hu, X., Li, J., Tang, J., and Yang, J. (2020). Generalized focal loss: Learning qualified and distributed bounding boxes for dense object detection. Advances in Neural Information Processing Systems (NeurIPS), Curran Associates, Inc."}],"container-title":["Journal of Imaging"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2313-433X\/12\/7\/325\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,21]],"date-time":"2026-07-21T04:23:12Z","timestamp":1784607792000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2313-433X\/12\/7\/325"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,18]]},"references-count":50,"journal-issue":{"issue":"7","published-online":{"date-parts":[[2026,7]]}},"alternative-id":["jimaging12070325"],"URL":"https:\/\/doi.org\/10.3390\/jimaging12070325","relation":{},"ISSN":["2313-433X"],"issn-type":[{"value":"2313-433X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,18]]}}}