{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/spatio-temporal-learnable-proposals-for-end","title":"Spatio-Temporal Learnable Proposals for End-to-End Video Object Detection","arxiv_id":"2210.02368","date":"2022-10-05","proceeding":null,"authors":["Khurram Azeem Hashmi","Didier Stricker","Muhammamd Zeshan Afzal"],"abstract":"This paper presents the novel idea of generating object proposals by leveraging temporal information for video object detection. The feature aggregation in modern region-based video object detectors heavily relies on learned proposals generated from a single-frame RPN. This imminently introduces additional components like NMS and produces unreliable proposals on low-quality frames. To tackle these restrictions, we present SparseVOD, a novel video object detection pipeline that employs Sparse R-CNN to exploit temporal information. In particular, we introduce two modules in the dynamic head of Sparse R-CNN. First, the Temporal Feature Extraction module based on the Temporal RoI Align operation is added to extract the RoI proposal features. Second, motivated by sequence-level semantic aggregation, we incorporate the attention-guided Semantic Proposal Feature Aggregation module to enhance object feature representation before detection. The proposed SparseVOD effectively alleviates the overhead of complicated post-processing methods and makes the overall pipeline end-to-end trainable. Extensive experiments show that our method significantly improves the single-frame Sparse RCNN by 8%-9% in mAP. Furthermore, besides achieving state-of-the-art 80.3% mAP on the ImageNet VID dataset with ResNet-50 backbone, our SparseVOD outperforms existing proposal-based methods by a significant margin on increasing IoU thresholds (IoU > 0.5).","url_abs":"https://arxiv.org/abs/2210.02368v2","url_pdf":"https://arxiv.org/pdf/2210.02368v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[],"tasks":[{"task_slug":"object","task_name":"Object"},{"task_slug":"object-detection","task_name":"Object Detection"},{"task_slug":"video-object-detection","task_name":"Video Object Detection"},{"task_slug":"object-detection-1","task_name":"object-detection"}],"methods":[{"method_slug":"align","method_name":"ALIGN"},{"method_slug":"rpn","method_name":"RPN"},{"method_slug":"sparse-r-cnn","method_name":"Sparse R-CNN"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/video-object-detection-on-imagenet-vid","task":"Video Object Detection","dataset":"ImageNet VID","model":"SparseVOD (ResNet-50)","rank_in_archive_order":27,"of":33,"metrics":{"MAP ":"80.3"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2210.02368","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}