Papers › Rethinking ImageNet Pre-training

Rethinking ImageNet Pre-training

21 Nov 2018ICCV 2019 10arXiv:1811.08883archive 2025-07-28

Kaiming He, Ross Girshick, Piotr Dollár

We report competitive results on object detection and instance segmentation on the COCO dataset using standard models trained from random initialization. The results are no worse than their ImageNet pre-training counterparts even when using the hyper-parameters of the baseline system (Mask R-CNN) that were optimized for fine-tuning pre-trained models, with the sole exception of increasing the number of training iterations so the randomly initialized models may converge. Training from random initialization is surprisingly robust; our results hold even when: (i) using only 10% of the training data, (ii) for deeper and wider models, and (iii) for multiple tasks and metrics. Experiments show that ImageNet pre-training speeds up convergence early in training, but does not necessarily provide regularization or improve final target task accuracy. To push the envelope we demonstrate 50.9 AP on COCO object detection without using any external data---a result on par with the top COCO 2017 competition results that used ImageNet pre-training. These observations challenge the conventional wisdom of ImageNet pre-training for dependent tasks and we expect these discoveries will encourage people to rethink the current de facto paradigm of `pre-training and fine-tuning' in computer vision.

PaperPDFConference PDFCode

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

Code

tensorpack/tensorpack mentioned on GitHubtf 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

Instance SegmentationObject DetectionSemantic Segmentationobject-detection

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Object Detection COCO minival Mask R-CNN (ResNeXt-152-FPN, cascade) AP50 66.8 #88 of 220 Archive leaderboard report
Object Detection COCO minival Mask R-CNN (ResNeXt-152-FPN, cascade) AP75 52.9 #88 of 220 Archive leaderboard report
Object Detection COCO minival Mask R-CNN (ResNeXt-152-FPN, cascade) box AP 48.6 #88 of 220 Archive leaderboard report
Object Detection COCO minival Mask R-CNN (ResNet-101-FPN, GN, Cascade) box AP 47.4 #96 of 220 Archive leaderboard report
Object Detection COCO minival Mask R-CNN (ResNeXt-152-FPN) AP50 67.1 #104 of 220 Archive leaderboard report
Object Detection COCO minival Mask R-CNN (ResNeXt-152-FPN) AP75 51.1 #104 of 220 Archive leaderboard report
Object Detection COCO minival Mask R-CNN (ResNeXt-152-FPN) box AP 46.4 #104 of 220 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

1x1 ConvolutionAverage PoolingBatch NormalizationBottleneck Residual BlockConvolutionGlobal Average PoolingGrouped ConvolutionKaiming InitializationMax PoolingReLUResNeXtResNeXt BlockResidual BlockResidual Connection

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