Papers › Is Heuristic Sampling Necessary in Training Deep Object Detectors?

Is Heuristic Sampling Necessary in Training Deep Object Detectors?

11 Sep 2019arXiv:1909.04868archive 2025-07-28

Joya Chen, Dong Liu, Tong Xu, Shiwei Wu, Yifei Cheng, Enhong Chen

To train accurate deep object detectors under the extreme foreground-background imbalance, heuristic sampling methods are always necessary, which either re-sample a subset of all training samples (hard sampling methods, e.g. biased sampling, OHEM), or use all training samples but re-weight them discriminatively (soft sampling methods, e.g. Focal Loss, GHM). In this paper, we challenge the necessity of such hard/soft sampling methods for training accurate deep object detectors. While previous studies have shown that training detectors without heuristic sampling methods would significantly degrade accuracy, we reveal that this degradation comes from an unreasonable classification gradient magnitude caused by the imbalance, rather than a lack of re-sampling/re-weighting. Motivated by our discovery, we propose a simple yet effective Sampling-Free mechanism to achieve a reasonable classification gradient magnitude by initialization and loss scaling. Unlike heuristic sampling methods with multiple hyperparameters, our Sampling-Free mechanism is fully data diagnostic, without laborious hyperparameters searching. We verify the effectiveness of our method in training anchor-based and anchor-free object detectors, where our method always achieves higher detection accuracy than heuristic sampling methods on COCO and PASCAL VOC datasets. Our Sampling-Free mechanism provides a new perspective to address the foreground-background imbalance. Our code is released at https://github.com/ChenJoya/sampling-free.

PaperPDFCode

Syntology This paper's arXiv id is not in Syntology's Atlas id list (549,449 ids, as this build read them; the read time is in the build record), so this page links to no Syntology page for it. The list is not the whole graph; Syntology's feed shows papers as they enter the graph.

Code

14 repositories listed; official and paper-mentioned ones first.

ChenJoya/objnessdet officialmentioned in papermentioned on GitHubpytorchSyntology: not harvested report
ChenJoya/sampling-free officialmentioned in papermentioned on GitHubpytorchSyntology: not harvested report
facebookresearch/maskrcnn-benchmark officialmentioned in papermentioned on GitHubpytorchMITSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Mask R-CNN (arXiv:1703.06870; that paper's own run record: 101 ran (of which 3 constructed an object rather than computing a result; 77 with no instrument failure: 2 honoured, 1 violated, 74 with no contract checked; 24 where Syntology's instrument failed) · 39 unverified); nothing checks that they implement this paper's method. report
FluteXu/ms-project mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to FCOS: Fully Convolutional One-Stage Object Detection (arXiv:1904.01355; that paper's own run record: 37 ran (of which 0 constructed an object rather than computing a result; 30 with no instrument failure: 2 honoured, 2 violated, 26 with no contract checked; 7 where Syntology's instrument failed) · 3 unverified); nothing checks that they implement this paper's method. report
GuoLiuFang/maskrcnn-benchmark-lfs mentioned on GitHubpytorchMITSyntology: not harvested report
SilvioGiancola/maskrcnn-benchmark mentioned on GitHubpytorchMITSyntology: not harvested report
Zhang-Jing-Xuan/MaskRCNN mentioned on GitHubpytorchnot reachable when probed 2026-09-18 — repositories for recent papers often appear after camera-readySyntology: not harvested report
banben/maskrcnn-benchmark mentioned on GitHubpytorchMITSyntology: not harvested report
basaltzhang/maskrcnn-benchmark mentioned on GitHubpytorchMITSyntology: not harvested report
ccchang1023/maskrcnn-benchmark mentioned on GitHubpytorchMITSyntology: not harvested report
cyctrung/DPnet mentioned on GitHubpytorchMITSyntology: not harvested report
jonvthvn90/Project mentioned on GitHubpytorchMITSyntology: not harvested report
markson14/WheatDet mentioned on GitHubpytorchMITSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to FCOS: Fully Convolutional One-Stage Object Detection (arXiv:1904.01355; that paper's own run record: 37 ran (of which 0 constructed an object rather than computing a result; 30 with no instrument failure: 2 honoured, 2 violated, 26 with no contract checked; 7 where Syntology's instrument failed) · 3 unverified); nothing checks that they implement this paper's method. report
ryota2425/maskrcnn-benchmark mentioned on GitHubpytorchMITSyntology: 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

DiagnosticGeneral ClassificationInstance SegmentationObjectSemantic Segmentation

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

Focal Loss

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