{"url":"/method/adaptive-feature-pooling","slug":"adaptive-feature-pooling","name":"Adaptive Feature Pooling","full_name":"Adaptive Feature Pooling","full_name_withheld":false,"description_markdown":"**Adaptive Feature Pooling** pools features from all levels for each proposal in object detection and fuses them for the following prediction. For each proposal, we map them to different feature levels. Following the idea of [Mask R-CNN](https://paperswithcode.com/method/adaptive-feature-pooling), [RoIAlign](https://paperswithcode.com/method/roi-align) is used to pool feature grids from each level. Then a fusion operation (element-wise max or sum) is utilized to fuse feature grids from different levels.\r\n\r\nThe motivation for this technique is that in an [FPN](https://paperswithcode.com/method/fpn) we assign proposals to different feature levels based on the size of proposals, which could be suboptimal if images with small differences are assigned to different levels, or if the importance of features is not strongly correlated to their level which they belong.","description_state":"present","introduced_year":null,"introduced_by":{"title":"Path Aggregation Network for Instance Segmentation","paper":"/paper/path-aggregation-network-for-instance","first_author":"Shu Liu","n_authors":5,"url_abs":null,"archive_paper_url":"https://paperswithcode.com/paper/path-aggregation-network-for-instance"},"source":{"url":"http://arxiv.org/abs/1803.01534v4","title":"Path Aggregation Network for Instance Segmentation","url_on_a_paper_host":true},"code_snippet_url":"https://github.com/ShuLiu1993/PANet/blob/2644d5ad6ae98c2bf58df45c8792c019b1d7b2b9/lib/model/roi_pooling/functions/roi_pool.py#L6","code_snippet_url_on_a_code_host":true,"categories":[{"area":"Computer Vision","area_id":"computer-vision","collection":"Pooling Operations","url":"/methods/category/pooling-operations","pwc_aliases":["pooling-operation"]}],"n_papers_tagged":20,"archive_num_papers":20,"papers_newest_first":[{"paper":null,"title":"YOLO-LLTS: Real-Time Low-Light Traffic Sign Detection via Prior-Guided Enhancement and Multi-Branch Feature Interaction","date":"2025-03-18","arxiv_id":"2503.13883","n_code_links":0,"syntology":null},{"paper":"/paper/a-lightweight-insulator-defect-detection","title":"A Lightweight Insulator Defect Detection Model Based on Drone Images","date":"2024-08-26","arxiv_id":null,"n_code_links":1,"syntology":null},{"paper":"/paper/csta-cnn-based-spatiotemporal-attention-for","title":"CSTA: CNN-based Spatiotemporal Attention for Video Summarization","date":"2024-05-20","arxiv_id":"2405.11905","n_code_links":1,"syntology":{"ran":7,"of":8,"unverified":1,"pointer_only":0}},{"paper":null,"title":"PANet: A Physics-guided Parametric Augmentation Net for Image Dehazing by Hazing","date":"2024-04-14","arxiv_id":"2404.09269","n_code_links":0,"syntology":null},{"paper":null,"title":"A Parallel Attention Network for Cattle Face Recognition","date":"2024-03-29","arxiv_id":"2403.19980","n_code_links":0,"syntology":null},{"paper":null,"title":"Toward Robust Canine Cardiac Diagnosis: Deep Prototype Alignment Network-Based Few-Shot Segmentation in Veterinary Medicine","date":"2024-03-11","arxiv_id":"2403.06471","n_code_links":0,"syntology":null},{"paper":null,"title":"More than the Sum of Its Parts: Ensembling Backbone Networks for Few-Shot Segmentation","date":"2024-02-09","arxiv_id":"2402.06581","n_code_links":0,"syntology":null},{"paper":"/paper/bridging-synthetic-and-real-worlds-for-pre","title":"Bridging Synthetic and Real Worlds for Pre-training Scene Text Detectors","date":"2023-12-08","arxiv_id":"2312.05286","n_code_links":1,"syntology":null},{"paper":null,"title":"Feature Aggregation in Joint Sound Classification and Localization Neural Networks","date":"2023-10-29","arxiv_id":"2310.19063","n_code_links":0,"syntology":null},{"paper":"/paper/panet-lidar-panoptic-segmentation-with-sparse","title":"PANet: LiDAR Panoptic Segmentation with Sparse Instance Proposal and Aggregation","date":"2023-06-27","arxiv_id":"2306.15348","n_code_links":1,"syntology":null},{"paper":null,"title":"Unconstrained Face Sketch Synthesis via Perception-Adaptive Network and A New Benchmark","date":"2021-12-02","arxiv_id":"2112.01019","n_code_links":0,"syntology":null},{"paper":null,"title":"PANet: Perspective-Aware Network with Dynamic Receptive Fields and Self-Distilling Supervision for Crowd Counting","date":"2021-10-31","arxiv_id":"2111.00406","n_code_links":0,"syntology":null},{"paper":null,"title":"CPNet: Cycle Prototype Network for Weakly-supervised 3D Renal Compartments Segmentation on CT Images","date":"2021-08-15","arxiv_id":"2108.06669","n_code_links":0,"syntology":null},{"paper":null,"title":"Weakly-supervised Part-Attention and Mentored Networks for Vehicle Re-Identification","date":"2021-07-17","arxiv_id":"2107.08228","n_code_links":0,"syntology":null},{"paper":"/paper/pyramid-attention-networks-for-image","title":"Pyramid Attention Networks for Image Restoration","date":"2020-04-28","arxiv_id":"2004.13824","n_code_links":2,"syntology":{"ran":1,"of":2,"unverified":1,"pointer_only":0}},{"paper":"/paper/yolov4-optimal-speed-and-accuracy-of-object","title":"YOLOv4: Optimal Speed and Accuracy of Object Detection","date":"2020-04-23","arxiv_id":"2004.10934","n_code_links":223,"syntology":{"ran":24,"of":184,"unverified":160,"pointer_only":8}},{"paper":"/paper/cspnet-a-new-backbone-that-can-enhance","title":"CSPNet: A New Backbone that can Enhance Learning Capability of CNN","date":"2019-11-27","arxiv_id":"1911.11929","n_code_links":123,"syntology":null},{"paper":"/paper/panet-few-shot-image-semantic-segmentation","title":"PANet: Few-Shot Image Semantic Segmentation with Prototype Alignment","date":"2019-08-18","arxiv_id":"1908.06391","n_code_links":5,"syntology":null},{"paper":"/paper/isaid-a-large-scale-dataset-for-instance","title":"iSAID: A Large-scale Dataset for Instance Segmentation in Aerial Images","date":"2019-05-30","arxiv_id":"1905.12886","n_code_links":3,"syntology":{"ran":1,"of":1,"unverified":0,"pointer_only":1}},{"paper":"/paper/path-aggregation-network-for-instance","title":"Path Aggregation Network for Instance Segmentation","date":"2018-03-05","arxiv_id":"1803.01534","n_code_links":10,"syntology":{"ran":1,"of":4,"unverified":3,"pointer_only":1}}],"papers_shown":20,"tasks":[{"task":"/task/segmentation","name":"Segmentation","papers":7},{"task":"/task/semantic-segmentation","name":"Semantic Segmentation","papers":6},{"task":"/task/object-detection","name":"Object Detection","papers":5},{"task":"/task/instance-segmentation","name":"Instance Segmentation","papers":3},{"task":"/task/object","name":"Object","papers":2},{"task":"/task/real-time-object-detection","name":"Real-Time Object Detection","papers":2},{"task":"/task/small-object-detection","name":"Small Object Detection","papers":2},{"task":"/task/object-detection-1","name":"object-detection","papers":2},{"task":"/task/anatomy","name":"Anatomy","papers":1},{"task":"/task/attribute","name":"Attribute","papers":1},{"task":"/task/autonomous-driving","name":"Autonomous Driving","papers":1},{"task":"/task/machine-learning","name":"BIG-bench Machine Learning","papers":1},{"task":"/task/crowd-counting","name":"Crowd Counting","papers":1},{"task":"/task/data-augmentation","name":"Data Augmentation","papers":1},{"task":"/task/decoder","name":"Decoder","papers":1},{"task":"/task/defect-detection","name":"Defect Detection","papers":1},{"task":"/task/demosaicking","name":"Demosaicking","papers":1},{"task":"/task/denoising","name":"Denoising","papers":1},{"task":"/task/face-recognition","name":"Face Recognition","papers":1},{"task":"/task/face-sketch-synthesis","name":"Face Sketch Synthesis","papers":1}],"tasks_shown":20,"n_tasks":44,"usage_by_year":[{"year":"2018","papers":1},{"year":"2019","papers":3},{"year":"2020","papers":2},{"year":"2021","papers":4},{"year":"2023","papers":3},{"year":"2024","papers":6},{"year":"2025","papers":1}],"row_source":"methods_table","archive":{"source":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","archive_url":"https://paperswithcode.com/method/adaptive-feature-pooling"},"syntology_read_at":"2026-09-24T18:15:14+00:00"}