{"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/pixel-wise-contrastive-distillation","title":"Pixel-Wise Contrastive Distillation","arxiv_id":"2211.00218","date":"2022-11-01","proceeding":"ICCV 2023 1","authors":["Junqiang Huang","Zichao Guo"],"abstract":"We present a simple but effective pixel-level self-supervised distillation framework friendly to dense prediction tasks. Our method, called Pixel-Wise Contrastive Distillation (PCD), distills knowledge by attracting the corresponding pixels from student's and teacher's output feature maps. PCD includes a novel design called SpatialAdaptor which ``reshapes'' a part of the teacher network while preserving the distribution of its output features. Our ablation experiments suggest that this reshaping behavior enables more informative pixel-to-pixel distillation. Moreover, we utilize a plug-in multi-head self-attention module that explicitly relates the pixels of student's feature maps to enhance the effective receptive field, leading to a more competitive student. PCD \\textbf{outperforms} previous self-supervised distillation methods on various dense prediction tasks. A backbone of \\mbox{ResNet-18-FPN} distilled by PCD achieves $37.4$ AP$^\\text{bbox}$ and $34.0$ AP$^\\text{mask}$ on COCO dataset using the detector of \\mbox{Mask R-CNN}. We hope our study will inspire future research on how to pre-train a small model friendly to dense prediction tasks in a self-supervised fashion.","url_abs":"https://arxiv.org/abs/2211.00218v3","url_pdf":"https://arxiv.org/pdf/2211.00218v3.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":"pixel-wise-contrastive-distillation","repo_url":"https://github.com/allo-rene/pcd","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"prediction","task_name":"Prediction"}],"methods":[{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"mask-r-cnn","method_name":"Mask R-CNN"},{"method_slug":"rpn","method_name":"RPN"},{"method_slug":"roi-align","method_name":"RoIAlign"},{"method_slug":"softmax","method_name":"Softmax"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2211.00218","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}