Papers › Class-Balanced Distillation for Long-Tailed Visual Recognition
Class-Balanced Distillation for Long-Tailed Visual Recognition
Ahmet Iscen, André Araujo, Boqing Gong, Cordelia Schmid
Real-world imagery is often characterized by a significant imbalance of the number of images per class, leading to long-tailed distributions. An effective and simple approach to long-tailed visual recognition is to learn feature representations and a classifier separately, with instance and class-balanced sampling, respectively. In this work, we introduce a new framework, by making the key observation that a feature representation learned with instance sampling is far from optimal in a long-tailed setting. Our main contribution is a new training method, referred to as Class-Balanced Distillation (CBD), that leverages knowledge distillation to enhance feature representations. CBD allows the feature representation to evolve in the second training stage, guided by the teacher learned in the first stage. The second stage uses class-balanced sampling, in order to focus on under-represented classes. This framework can naturally accommodate the usage of multiple teachers, unlocking the information from an ensemble of models to enhance recognition capabilities. Our experiments show that the proposed technique consistently outperforms the state of the art on long-tailed recognition benchmarks such as ImageNet-LT, iNaturalist17 and iNaturalist18.
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):
get_citation_path(paper_1="2104.05279", paper_2="…")with another paper's arXiv id or titleget_concepts_for_paper(arxiv_id="2104.05279")
Code
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
Results from the paper archive 2025-07-28
| Task | Dataset | Model | Metric | Value | Rank at snapshot | Leaderboard | Report |
|---|---|---|---|---|---|---|---|
| Image Classification | iNaturalist 2018 | CBD-ENS (ResNet-101) | Top-1 Accuracy | 75.3% | #26 of 60 | Archive leaderboard | report |
| Image Classification | iNaturalist 2018 | CBD-ENS (ResNet-50) | Top-1 Accuracy | 73.6% | #31 of 60 | Archive leaderboard | report |
| Long-tail Learning | ImageNet-LT | CBD-ENS (ResNet-152) | Top-1 Accuracy | 57.7 | #26 of 69 | Archive leaderboard | report |
| Long-tail Learning | ImageNet-LT | CBD-ENS (ResNet-50) | Top-1 Accuracy | 55.6 | #36 of 69 | Archive leaderboard | report |
| Long-tail Learning | iNaturalist 2018 | CBD-ENS (ResNet-101) | Top-1 Accuracy | 75.3% | #13 of 43 | Archive leaderboard | report |
| Long-tail Learning | iNaturalist 2018 | CBD-ENS (ResNet-50) | Top-1 Accuracy | 73.6% | #21 of 43 | 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
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