{"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/carafe-content-aware-reassembly-of-features","title":"CARAFE: Content-Aware ReAssembly of FEatures","arxiv_id":"1905.02188","date":"2019-05-06","proceeding":"ICCV 2019 10","authors":["Jiaqi Wang","Kai Chen","Rui Xu","Ziwei Liu","Chen Change Loy","Dahua Lin"],"abstract":"Feature upsampling is a key operation in a number of modern convolutional network architectures, e.g. feature pyramids. Its design is critical for dense prediction tasks such as object detection and semantic/instance segmentation. In this work, we propose Content-Aware ReAssembly of FEatures (CARAFE), a universal, lightweight and highly effective operator to fulfill this goal. CARAFE has several appealing properties: (1) Large field of view. Unlike previous works (e.g. bilinear interpolation) that only exploit sub-pixel neighborhood, CARAFE can aggregate contextual information within a large receptive field. (2) Content-aware handling. Instead of using a fixed kernel for all samples (e.g. deconvolution), CARAFE enables instance-specific content-aware handling, which generates adaptive kernels on-the-fly. (3) Lightweight and fast to compute. CARAFE introduces little computational overhead and can be readily integrated into modern network architectures. We conduct comprehensive evaluations on standard benchmarks in object detection, instance/semantic segmentation and inpainting. CARAFE shows consistent and substantial gains across all the tasks (1.2%, 1.3%, 1.8%, 1.1db respectively) with negligible computational overhead. It has great potential to serve as a strong building block for future research. It has great potential to serve as a strong building block for future research. Code and models are available at https://github.com/open-mmlab/mmdetection.","url_abs":"https://arxiv.org/abs/1905.02188v3","url_pdf":"https://arxiv.org/pdf/1905.02188v3.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":"carafe-content-aware-reassembly-of-features","repo_url":"https://github.com/open-mmlab/mmdetection","is_official":0,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"carafe-content-aware-reassembly-of-features","repo_url":"https://github.com/smallsunsun1/custom_ops","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}},{"paper_slug":"carafe-content-aware-reassembly-of-features","repo_url":"https://github.com/2023-MindSpore-1/ms-code-16/tree/main/Contents","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null}],"tasks":[{"task_slug":"feature-upsampling","task_name":"Feature Upsampling"},{"task_slug":"instance-segmentation","task_name":"Instance Segmentation"},{"task_slug":"object-detection","task_name":"Object Detection"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"},{"task_slug":"object-detection-1","task_name":"object-detection"}],"methods":[{"method_slug":"carafe","method_name":"CARAFE"}],"datasets_introduced":[],"methods_introduced":[{"slug":"carafe","name":"CARAFE","full_name":"CARAFE"}],"results":[{"leaderboard":"/sota/feature-upsampling-on-imagenet","task":"Feature Upsampling","dataset":"ImageNet","model":"CARAFE","rank_in_archive_order":5,"of":8,"metrics":{"ADCC":"59.7","Average Drop":"49.9","Average Increase":"4.8"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/1905.02188","atlas_url":"https://app.syntology.ai/?focus=1905.02188","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}