Papers › Repulsion Loss: Detecting Pedestrians in a Crowd

Repulsion Loss: Detecting Pedestrians in a Crowd

21 Nov 2017CVPR 2018 6arXiv:1711.07752archive 2025-07-28

Xinlong Wang, Tete Xiao, Yuning Jiang, Shuai Shao, Jian Sun, Chunhua Shen

Detecting individual pedestrians in a crowd remains a challenging problem since the pedestrians often gather together and occlude each other in real-world scenarios. In this paper, we first explore how a state-of-the-art pedestrian detector is harmed by crowd occlusion via experimentation, providing insights into the crowd occlusion problem. Then, we propose a novel bounding box regression loss specifically designed for crowd scenes, termed repulsion loss. This loss is driven by two motivations: the attraction by target, and the repulsion by other surrounding objects. The repulsion term prevents the proposal from shifting to surrounding objects thus leading to more crowd-robust localization. Our detector trained by repulsion loss outperforms all the state-of-the-art methods with a significant improvement in occlusion cases.

PaperPDFConference PDFCode

In Syntology View this paper on Syntology, its page in Syntology's graph. That page lists the repositories linked to the paper, the abstract and the calls for agents.

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

For agents, on Syntology's MCP service (how to connect):

Code

bailvwangzi/repulsion_loss_ssd mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to SSD: Single Shot MultiBox Detector (arXiv:1512.02325; that paper's own run record: 93 ran (of which 0 constructed an object rather than computing a result; 78 with no instrument failure: 3 honoured, 1 violated, 74 with no contract checked; 15 where Syntology's instrument failed) · 38 unverified); nothing checks that they implement this paper's method. report
justinkay/repulsion-loss-detectron2 mentioned on GitHubpytorchSyntology: not harvested 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.

A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.

Code Syntology ran Syntology

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

Tasks

Pedestrian Detectionregression

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Pedestrian Detection Caltech RepLoss + CityPersons dataset Reasonable Miss Rate 4.0 #9 of 33 Archive leaderboard report
Pedestrian Detection Caltech RepLoss Reasonable Miss Rate 5.0 #13 of 33 Archive leaderboard report
Pedestrian Detection CityPersons RepLoss Bare MR^-2 7.6 #17 of 22 Archive leaderboard report
Pedestrian Detection CityPersons RepLoss Heavy MR^-2 56.9 #17 of 22 Archive leaderboard report
Pedestrian Detection CityPersons RepLoss Partial MR^-2 16.8 #17 of 22 Archive leaderboard report
Pedestrian Detection CityPersons RepLoss Reasonable MR^-2 13.2 #17 of 22 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