AI Paper Atlas

Research map / Benchmarks / Birdsnap

Birdsnap leaderboard

10 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on Birdsnap. Best accuracy: 85.4 (L-Bird (MC), ECCV 2016).

Numbers are as each paper reports them, read by AI from its results tables and not checked by hand. Papers can differ in setting (training data, model size, extra data); the topic column shows what each paper is about. Open the paper before you compare or quote a number.

Birdsnap · accuracy

10 papers, each with its best reported number for its own method; higher is better.

#accuracy ↑MethodPaperTopic
185.4L-Bird (MC)
Inception-v3
The Unreasonable Effectiveness of Noisy Data for Fine-Grained Recognition
ECCV 2016 · Table 1
Recognition and long-tail
284.3SENet-154 + our method
SENet-154
Fixing the train-test resolution discrepancy
NeurIPS 2019 · Table 3
Test-time adaptation
383.71K-MIM
SwinV2-B
Revealing the Dark Secrets of Masked Image Modeling
CVPR 2023 · Appendix: Detailed Results on
Attention and transformers
483.6Full (JFT)
ResNet50-v2
Scalable Transfer Learning with Expert Models
ICLR 2021 · Natural datasets comparison ta
Recognition and long-tail
583.6AmoebaNet-B (18, 512) fine-tuned from ImageNet
AmoebaNet-B (18, 512)
GPipe: Efficient Training of Giant Neural Networks using Pipeline Parallelism
NeurIPS 2019 · Table: Image classification ac
RNNs and normalization
679.44Space2Vec gridMulti-Scale Representation Learning for Spatial Feature Distributions using Grid Cells
ICLR 2020 · Table 4
Recognition and long-tail
778.4Inception v4, fine-tuned
Inception v4
Do Better ImageNet Models Transfer Better?
CVPR 2019 · Appendix, Comparison versus st
CNN design
876.41Robust ImageNet model, full-network
ImageNet classifier
Do Adversarially Robust ImageNet Models Transfer Better?
NeurIPS 2020 · Table: transfer performance
Recognition and long-tail
955.35SPU
CLIP-ViT/B-16
Overcoming Generic Knowledge Loss with Selective Parameter Update
CVPR 2024 · Table 1
Continual fine-tuning of LLMs
1053.46t-MoCo-v2
MoCo-v2
Your Contrastive Learning Is Secretly Doing Stochastic Neighbor Embedding
ICLR 2023 · Table 1
Self-supervised and contrastive

Not affiliated with any conference, journal or the benchmark's authors. Corrections: contact@latenticlabs.com.