Papers › F2DNet: Fast Focal Detection Network for Pedestrian Detection

F2DNet: Fast Focal Detection Network for Pedestrian Detection

4 Mar 2022arXiv:2203.02331archive 2025-07-28

Abdul Hannan Khan, Mohsin Munir, Ludger van Elst, Andreas Dengel

Two-stage detectors are state-of-the-art in object detection as well as pedestrian detection. However, the current two-stage detectors are inefficient as they do bounding box regression in multiple steps i.e. in region proposal networks and bounding box heads. Also, the anchor-based region proposal networks are computationally expensive to train. We propose F2DNet, a novel two-stage detection architecture which eliminates redundancy of current two-stage detectors by replacing the region proposal network with our focal detection network and bounding box head with our fast suppression head. We benchmark F2DNet on top pedestrian detection datasets, thoroughly compare it against the existing state-of-the-art detectors and conduct cross dataset evaluation to test the generalizability of our model to unseen data. Our F2DNet achieves 8.7\%, 2.2\%, and 6.1\% MR-2 on City Persons, Caltech Pedestrian, and Euro City Person datasets respectively when trained on a single dataset and reaches 20.4\% and 26.2\% MR-2 in heavy occlusion setting of Caltech Pedestrian and City Persons datasets when using progressive fine-tunning. Furthermore, F2DNet have significantly lesser inference time compared to the current state-of-the-art. Code and trained models will be available at https://github.com/AbdulHannanKhan/F2DNet.

PaperPDFCode

Code

abdulhannankhan/f2dnet officialmentioned in papermentioned on GitHubpytorch report
hasanirtiza/Pedestron mentioned on GitHubpytorchApache-2.0 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

Object DetectionPedestrian DetectionRegion Proposalobject-detection

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Pedestrian Detection Caltech F2DNet (extra data) Heavy MR^-2 20.42 #2 of 33 Archive leaderboard report
Pedestrian Detection Caltech F2DNet (extra data) Reasonable Miss Rate 1.71 #2 of 33 Archive leaderboard report
Pedestrian Detection Caltech F2DNet Heavy MR^-2 38.7 #4 of 33 Archive leaderboard report
Pedestrian Detection Caltech F2DNet Reasonable Miss Rate 2.2 #4 of 33 Archive leaderboard report
Pedestrian Detection CityPersons F2DNet (extra data) Heavy MR^-2 26.23 #4 of 22 Archive leaderboard report
Pedestrian Detection CityPersons F2DNet (extra data) Reasonable MR^-2 7.8 #4 of 22 Archive leaderboard report
Pedestrian Detection CityPersons F2DNet (extra data) Small MR^-2 9.43 #4 of 22 Archive leaderboard report
Pedestrian Detection CityPersons F2DNet (extra data) Test Time 0.44s/img #4 of 22 Archive leaderboard report
Pedestrian Detection CityPersons F2DNet Heavy MR^-2 32.6 #6 of 22 Archive leaderboard report
Pedestrian Detection CityPersons F2DNet Reasonable MR^-2 8.7 #6 of 22 Archive leaderboard report
Pedestrian Detection CityPersons F2DNet Small MR^-2 11.3 #6 of 22 Archive leaderboard report
Pedestrian Detection CityPersons F2DNet Test Time 0.44s/img #6 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.

Methods

Introduced by this paper: F2DNet

F2DNet

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