Papers › Location-Sensitive Visual Recognition with Cross-IOU Loss

Location-Sensitive Visual Recognition with Cross-IOU Loss

11 Apr 2021arXiv:2104.04899archive 2025-07-28

Kaiwen Duan, Lingxi Xie, Honggang Qi, Song Bai, Qingming Huang, Qi Tian

Object detection, instance segmentation, and pose estimation are popular visual recognition tasks which require localizing the object by internal or boundary landmarks. This paper summarizes these tasks as location-sensitive visual recognition and proposes a unified solution named location-sensitive network (LSNet). Based on a deep neural network as the backbone, LSNet predicts an anchor point and a set of landmarks which together define the shape of the target object. The key to optimizing the LSNet lies in the ability of fitting various scales, for which we design a novel loss function named cross-IOU loss that computes the cross-IOU of each anchor point-landmark pair to approximate the global IOU between the prediction and ground-truth. The flexibly located and accurately predicted landmarks also enable LSNet to incorporate richer contextual information for visual recognition. Evaluated on the MS-COCO dataset, LSNet set the new state-of-the-art accuracy for anchor-free object detection (a 53.5% box AP) and instance segmentation (a 40.2% mask AP), and shows promising performance in detecting multi-scale human poses. Code is available at https://github.com/Duankaiwen/LSNet

PaperPDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

Duankaiwen/LSNet officialmentioned in papermentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

2D Human Pose EstimationInstance SegmentationObjectObject DetectionPose EstimationSemantic Segmentationobject-detection

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Object Detection COCO test-dev LSNet (Res2Net-101+ DCN, multi-scale) AP50 71.1 #61 of 225 Archive leaderboard report
Object Detection COCO test-dev LSNet (Res2Net-101+ DCN, multi-scale) AP75 59.2 #61 of 225 Archive leaderboard report
Object Detection COCO test-dev LSNet (Res2Net-101+ DCN, multi-scale) APL 65.8 #61 of 225 Archive leaderboard report
Object Detection COCO test-dev LSNet (Res2Net-101+ DCN, multi-scale) APM 56.4 #61 of 225 Archive leaderboard report
Object Detection COCO test-dev LSNet (Res2Net-101+ DCN, multi-scale) APS 35.2 #61 of 225 Archive leaderboard report
Object Detection COCO test-dev LSNet (Res2Net-101+ DCN, multi-scale) box mAP 53.5 #61 of 225 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections