{"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/clipself-vision-transformer-distills-itself","title":"CLIPSelf: Vision Transformer Distills Itself for Open-Vocabulary Dense Prediction","arxiv_id":"2310.01403","date":"2023-10-02","proceeding":null,"authors":["Size Wu","Wenwei Zhang","Lumin Xu","Sheng Jin","Xiangtai Li","Wentao Liu","Chen Change Loy"],"abstract":"Open-vocabulary dense prediction tasks including object detection and image segmentation have been advanced by the success of Contrastive Language-Image Pre-training (CLIP). CLIP models, particularly those incorporating vision transformers (ViTs), have exhibited remarkable generalization ability in zero-shot image classification. However, when transferring the vision-language alignment of CLIP from global image representation to local region representation for the open-vocabulary dense prediction tasks, CLIP ViTs suffer from the domain shift from full images to local image regions. In this paper, we embark on an in-depth analysis of the region-language alignment in CLIP models, which is essential for downstream open-vocabulary dense prediction tasks. Subsequently, we propose an approach named CLIPSelf, which adapts the image-level recognition ability of CLIP ViT to local image regions without needing any region-text pairs. CLIPSelf empowers ViTs to distill itself by aligning a region representation extracted from its dense feature map with the image-level representation of the corresponding image crop. With the enhanced CLIP ViTs, we achieve new state-of-the-art performance on open-vocabulary object detection, semantic segmentation, and panoptic segmentation across various benchmarks. Models and code are released at https://github.com/wusize/CLIPSelf.","url_abs":"https://arxiv.org/abs/2310.01403v2","url_pdf":"https://arxiv.org/pdf/2310.01403v2.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":"clipself-vision-transformer-distills-itself","repo_url":"https://github.com/wusize/clipself","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"NOASSERTION"}}],"tasks":[{"task_slug":"image-classification","task_name":"Image Classification"},{"task_slug":"image-segmentation","task_name":"Image Segmentation"},{"task_slug":"object-detection","task_name":"Object Detection"},{"task_slug":"open-vocabulary-object-detection","task_name":"Open Vocabulary Object Detection"},{"task_slug":"open-vocabulary-panoptic-segmentation","task_name":"Open Vocabulary Panoptic Segmentation"},{"task_slug":"open-vocabulary-semantic-segmentation","task_name":"Open Vocabulary Semantic Segmentation"},{"task_slug":"open-vocabulary-object-detection","task_name":"Open-vocabulary object detection"},{"task_slug":"panoptic-segmentation","task_name":"Panoptic Segmentation"},{"task_slug":"prediction","task_name":"Prediction"},{"task_slug":"segmentation","task_name":"Segmentation"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"},{"task_slug":"zero-shot-image-classification","task_name":"Zero-Shot Image Classification"},{"task_slug":"image-classification","task_name":"image-classification"},{"task_slug":"object-detection-1","task_name":"object-detection"}],"methods":[{"method_slug":"clip","method_name":"CLIP"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/open-vocabulary-object-detection-on-lvis-v1-0","task":"Open Vocabulary Object Detection","dataset":"LVIS v1.0","model":"CLIPSelf","rank_in_archive_order":5,"of":28,"metrics":{"AP novel-LVIS base training":"34.9"},"uses_additional_data":false},{"leaderboard":"/sota/open-vocabulary-object-detection-on-mscoco","task":"Open Vocabulary Object Detection","dataset":"MSCOCO","model":"CLIPSelf","rank_in_archive_order":7,"of":32,"metrics":{"AP 0.5":"44.3"},"uses_additional_data":false},{"leaderboard":"/sota/open-vocabulary-panoptic-segmentation-on","task":"Open Vocabulary Panoptic Segmentation","dataset":"ADE20K","model":"CLIPSelf","rank_in_archive_order":6,"of":10,"metrics":{"PQ":"23.7"},"uses_additional_data":false},{"leaderboard":"/sota/open-vocabulary-semantic-segmentation-on-2","task":"Open Vocabulary Semantic Segmentation","dataset":"ADE20K-150","model":"CLIPSelf","rank_in_archive_order":10,"of":23,"metrics":{"mIoU":"34.5"},"uses_additional_data":false},{"leaderboard":"/sota/open-vocabulary-semantic-segmentation-on-3","task":"Open Vocabulary Semantic Segmentation","dataset":"ADE20K-847","model":"CLIPSelf","rank_in_archive_order":14,"of":19,"metrics":{"mIoU":"12.4"},"uses_additional_data":false},{"leaderboard":"/sota/open-vocabulary-semantic-segmentation-on-1","task":"Open Vocabulary Semantic Segmentation","dataset":"PASCAL Context-59","model":"CLIPSelf","rank_in_archive_order":5,"of":24,"metrics":{"mIoU":"62.3"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2310.01403","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2310.01403"}},"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/wusize/clipself","reach":{"status":"ok","spdx":"NOASSERTION"}}],"summary":{"ran":2,"unverified":4},"by_repo_kind":{"official":{"samples":6,"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":6,"samples":[{"code_sha256_prefix":"42264faf65871374","entry":"load_class_freq","repo":"wusize/clipself","repo_kind":"official","path":"F-ViT/models/custom_losses.py","file_url":"https://github.com/wusize/clipself/blob/HEAD/F-ViT/models/custom_losses.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"42264faf65871374"}},{"code_sha256_prefix":"4b3a4d17e1df3140","entry":"load_state_dict","repo":"wusize/clipself","repo_kind":"official","path":"src/open_clip/factory.py","file_url":"https://github.com/wusize/clipself/blob/HEAD/src/open_clip/factory.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"4b3a4d17e1df3140"}},{"code_sha256_prefix":"fd6321f5fd69ea0a","entry":"gather_features","repo":"wusize/clipself","repo_kind":"official","path":"src/open_clip/loss.py","file_url":"https://github.com/wusize/clipself/blob/HEAD/src/open_clip/loss.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"fd6321f5fd69ea0a"}},{"code_sha256_prefix":"5c7cf820e99443ff","entry":"get_model_config","repo":"wusize/clipself","repo_kind":"official","path":"src/open_clip/factory.py","file_url":"https://github.com/wusize/clipself/blob/HEAD/src/open_clip/factory.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"5c7cf820e99443ff"}},{"code_sha256_prefix":"fb651d0a97fd4d3f","entry":"prepare_inputs_for_generation","repo":"wusize/clipself","repo_kind":"official","path":"src/open_clip/coca_model.py","file_url":"https://github.com/wusize/clipself/blob/HEAD/src/open_clip/coca_model.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"fb651d0a97fd4d3f"}},{"code_sha256_prefix":"2a377da4a76a2d44","entry":"register_pooler","repo":"wusize/clipself","repo_kind":"official","path":"src/open_clip/hf_model.py","file_url":"https://github.com/wusize/clipself/blob/HEAD/src/open_clip/hf_model.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"2a377da4a76a2d44"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}