</> Code with Papers by Syntology

Papers › Grafit: Learning fine-grained image representations with coarse labels

Grafit: Learning fine-grained image representations with coarse labels

25 Nov 2020ICCV 2021 10arXiv:2011.12982archive 2025-07-28

Hugo Touvron, Alexandre Sablayrolles, Matthijs Douze, Matthieu Cord, Hervé Jégou

This paper tackles the problem of learning a finer representation than the one provided by training labels. This enables fine-grained category retrieval of images in a collection annotated with coarse labels only. Our network is learned with a nearest-neighbor classifier objective, and an instance loss inspired by self-supervised learning. By jointly leveraging the coarse labels and the underlying fine-grained latent space, it significantly improves the accuracy of category-level retrieval methods. Our strategy outperforms all competing methods for retrieving or classifying images at a finer granularity than that available at train time. It also improves the accuracy for transfer learning tasks to fine-grained datasets, thereby establishing the new state of the art on five public benchmarks, like iNaturalist-2018.

PaperPDFConference PDF

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

Code

No code repository is listed for this paper in the archive or in Syntology's graph.

Code Syntology ran Syntology

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

Tasks

Fine-Grained Image ClassificationImage ClassificationLearning with coarse labelsRetrievalSelf-Supervised LearningTransfer Learning

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Fine-Grained Image Classification Food-101 Grafit (RegNet-8GF) Accuracy 93.7 #6 of 15 Archive leaderboard report
Fine-Grained Image Classification Oxford 102 Flowers Grafit (RegNet-8GF) Accuracy 99.1% #7 of 25 Archive leaderboard report
Fine-Grained Image Classification Stanford Cars Grafit (RegNet-8GF) Accuracy 94.7% #36 of 83 Archive leaderboard report
Image Classification CIFAR-100 Grafit (ResNet-50) Percentage correct 83.7 #84 of 211 Archive leaderboard report
Image Classification Flowers-102 Grafit (RegNet-8GF) Accuracy 99.1% #15 of 52 Archive leaderboard report
Image Classification ImageNet Grafit (ResNet-50) Top 1 Accuracy 79.6% #747 of 1060 Archive leaderboard report
Image Classification iNaturalist 2018 RegNet-8GF Top-1 Accuracy 81.2% #13 of 60 Archive leaderboard report
Image Classification iNaturalist 2018 ResNet-50 Top-1 Accuracy 69.8% #39 of 60 Archive leaderboard report
Image Classification iNaturalist 2019 Grafit (RegnetY 8GF) Top-1 Accuracy 84.1 #3 of 22 Archive leaderboard report
Learning with coarse labels ImageNet32 Grafit Recall@1 18.13 #2 of 2 Archive leaderboard report
Learning with coarse labels ImageNet32 Grafit Recall@10 46.64 #2 of 2 Archive leaderboard report
Learning with coarse labels ImageNet32 Grafit Recall@2 25.46 #2 of 2 Archive leaderboard report
Learning with coarse labels ImageNet32 Grafit Recall@5 37.19 #2 of 2 Archive leaderboard report
Learning with coarse labels Stanford Cars Grafit Recall@1 42.30 #2 of 2 Archive leaderboard report
Learning with coarse labels Stanford Cars Grafit Recall@10 81.74 #2 of 2 Archive leaderboard report
Learning with coarse labels Stanford Cars Grafit Recall@2 54.79 #2 of 2 Archive leaderboard report
Learning with coarse labels Stanford Cars Grafit Recall@5 71.1 #2 of 2 Archive leaderboard report
Learning with coarse labels Stanford Online Products Grafit Recall@1 74.02 #2 of 2 Archive leaderboard report
Learning with coarse labels Stanford Online Products Grafit Recall@10 87.91 #2 of 2 Archive leaderboard report
Learning with coarse labels Stanford Online Products Grafit Recall@2 78.82 #2 of 2 Archive leaderboard report
Learning with coarse labels Stanford Online Products Grafit Recall@5 84.13 #2 of 2 Archive leaderboard report
Learning with coarse labels cifar100 Grafit Recall@1 60.57 #2 of 2 Archive leaderboard report
Learning with coarse labels cifar100 Grafit Recall@10 89.21 #2 of 2 Archive leaderboard report
Learning with coarse labels cifar100 Grafit Recall@2 71.13 #2 of 2 Archive leaderboard report
Learning with coarse labels cifar100 Grafit Recall@5 82.32 #2 of 2 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