{"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/cbnetv2-a-composite-backbone-network","title":"CBNet: A Composite Backbone Network Architecture for Object Detection","arxiv_id":"2107.00420","date":"2021-07-01","proceeding":null,"authors":["TingTing Liang","Xiaojie Chu","Yudong Liu","Yongtao Wang","Zhi Tang","Wei Chu","Jingdong Chen","Haibin Ling"],"abstract":"Modern top-performing object detectors depend heavily on backbone networks, whose advances bring consistent performance gains through exploring more effective network structures. In this paper, we propose a novel and flexible backbone framework, namely CBNetV2, to construct high-performance detectors using existing open-sourced pre-trained backbones under the pre-training fine-tuning paradigm. In particular, CBNetV2 architecture groups multiple identical backbones, which are connected through composite connections. Specifically, it integrates the high- and low-level features of multiple backbone networks and gradually expands the receptive field to more efficiently perform object detection. We also propose a better training strategy with assistant supervision for CBNet-based detectors. Without additional pre-training of the composite backbone, CBNetV2 can be adapted to various backbones (CNN-based vs. Transformer-based) and head designs of most mainstream detectors (one-stage vs. two-stage, anchor-based vs. anchor-free-based). Experiments provide strong evidence that, compared with simply increasing the depth and width of the network, CBNetV2 introduces a more efficient, effective, and resource-friendly way to build high-performance backbone networks. Particularly, our Dual-Swin-L achieves 59.4% box AP and 51.6% mask AP on COCO test-dev under the single-model and single-scale testing protocol, which is significantly better than the state-of-the-art result (57.7% box AP and 50.2% mask AP) achieved by Swin-L, while the training schedule is reduced by 6$\\times$. With multi-scale testing, we push the current best single model result to a new record of 60.1% box AP and 52.3% mask AP without using extra training data. Code is available at https://github.com/VDIGPKU/CBNetV2.","url_abs":"https://arxiv.org/abs/2107.00420v7","url_pdf":"https://arxiv.org/pdf/2107.00420v7.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":[{"paper_slug":"cbnetv2-a-composite-backbone-network","repo_url":"https://github.com/VDIGPKU/CBNetV2","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"cbnetv2-a-composite-backbone-network","repo_url":"https://github.com/epsilon-deltta/epsilon-deltta","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}},{"paper_slug":"cbnetv2-a-composite-backbone-network","repo_url":"https://github.com/PaddlePaddle/PaddleDetection","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"paddle","reach":null},{"paper_slug":"cbnetv2-a-composite-backbone-network","repo_url":"https://github.com/shinya7y/UniverseNet","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"instance-segmentation","task_name":"Instance Segmentation"},{"task_slug":"object","task_name":"Object"},{"task_slug":"object-detection","task_name":"Object Detection"},{"task_slug":"real-time-object-detection","task_name":"Real-Time Object Detection"},{"task_slug":"object-detection-1","task_name":"object-detection"}],"methods":[{"method_slug":"1x1-convolution","method_name":"1x1 Convolution"},{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"fpn","method_name":"FPN"},{"method_slug":"htc","method_name":"HTC"},{"method_slug":"rpn","method_name":"RPN"},{"method_slug":"roi-align","method_name":"RoIAlign"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/instance-segmentation-on-coco-minival","task":"Instance Segmentation","dataset":"COCO minival","model":"CBNetV2 (Dual-Swin-L HTC, multi-scale)","rank_in_archive_order":18,"of":93,"metrics":{"mask AP":"51.8"},"uses_additional_data":false},{"leaderboard":"/sota/instance-segmentation-on-coco-minival","task":"Instance Segmentation","dataset":"COCO minival","model":"CBNetV2 (Dual-Swin-L HTC, multi-scale)","rank_in_archive_order":20,"of":93,"metrics":{"mask AP":"51"},"uses_additional_data":false},{"leaderboard":"/sota/instance-segmentation-on-coco","task":"Instance Segmentation","dataset":"COCO test-dev","model":"CBNetV2 (EVA02, single-scale)","rank_in_archive_order":2,"of":112,"metrics":{"AP50":"80.3","AP75":"62.1","APL":"70.9","APM":"59.3","APS":"39.7","mask AP":"56.1"},"uses_additional_data":true},{"leaderboard":"/sota/instance-segmentation-on-coco","task":"Instance Segmentation","dataset":"COCO test-dev","model":"CBNetV2 (Dual-Swin-L HTC, multi-scale)","rank_in_archive_order":16,"of":112,"metrics":{"mask AP":"52.3"},"uses_additional_data":false},{"leaderboard":"/sota/instance-segmentation-on-coco","task":"Instance Segmentation","dataset":"COCO test-dev","model":"CBNetV2 (Dual-Swin-L HTC, single-scale)","rank_in_archive_order":18,"of":112,"metrics":{"mask AP":"51.6"},"uses_additional_data":false},{"leaderboard":"/sota/object-detection-on-coco-minival","task":"Object Detection","dataset":"COCO minival","model":"CBNetV2 (Dual-Swin-L HTC, multi-scale)","rank_in_archive_order":28,"of":220,"metrics":{"box AP":"59.6"},"uses_additional_data":false},{"leaderboard":"/sota/object-detection-on-coco-minival","task":"Object Detection","dataset":"COCO minival","model":"CBNetV2 (Dual-Swin-L HTC, multi-scale)","rank_in_archive_order":32,"of":220,"metrics":{"box AP":"59.1"},"uses_additional_data":false},{"leaderboard":"/sota/object-detection-on-coco","task":"Object Detection","dataset":"COCO test-dev","model":"CBNetV2 (Dual-Swin-L HTC, multi-scale)","rank_in_archive_order":30,"of":225,"metrics":{"box mAP":"60.1"},"uses_additional_data":false},{"leaderboard":"/sota/object-detection-on-coco","task":"Object Detection","dataset":"COCO test-dev","model":"CBNetV2 (Dual-Swin-L HTC, single-scale)","rank_in_archive_order":32,"of":225,"metrics":{"box mAP":"59.4"},"uses_additional_data":false},{"leaderboard":"/sota/object-detection-on-coco-o","task":"Object Detection","dataset":"COCO-O","model":"CBNetV2\n(Swin-L)","rank_in_archive_order":6,"of":45,"metrics":{"Average mAP":"39.0","Effective Robustness":"12.36"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2107.00420","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2107.00420"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/epsilon-deltta/epsilon-deltta","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/VDIGPKU/CBNetV2","reach":{"status":"ok","spdx":"Apache-2.0"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/PaddlePaddle/PaddleDetection","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/shinya7y/UniverseNet","reach":{"status":"ok","spdx":"Apache-2.0"}}],"summary":{"ran":2,"unverified":1},"by_repo_kind":{"official":{"samples":3,"ran":2,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"ae35528c6bd4bd1c","entry":"process_checkpoint","repo":"VDIGPKU/CBNetV2","repo_kind":"official","path":".dev_scripts/gather_models.py","file_url":"https://github.com/VDIGPKU/CBNetV2/blob/HEAD/.dev_scripts/gather_models.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"ae35528c6bd4bd1c"}},{"code_sha256_prefix":"893e052530d28946","entry":"process_model_info","repo":"VDIGPKU/CBNetV2","repo_kind":"official","path":".dev_scripts/convert_test_benchmark_script.py","file_url":"https://github.com/VDIGPKU/CBNetV2/blob/HEAD/.dev_scripts/convert_test_benchmark_script.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"893e052530d28946"}},{"code_sha256_prefix":"8ff99a6bd2ebf598","entry":"ordered_yaml_dump","repo":"VDIGPKU/CBNetV2","repo_kind":"official","path":".dev_scripts/gather_models.py","file_url":"https://github.com/VDIGPKU/CBNetV2/blob/HEAD/.dev_scripts/gather_models.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"8ff99a6bd2ebf598"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}