AI Paper Atlas

Research map / Benchmarks / CUB 5-shot

CUB 5-shot leaderboard

17 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on CUB 5-shot. Best accuracy: 95.96 (OT-GNN, TPAMI 2023).

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.

CUB 5-shot · accuracy

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

#accuracy ↑MethodPaperTopic
195.96OT-GNN
ResNet-18
Omni-Training: Bridging Pre-Training and Meta-Training for Few-Shot Learning
TPAMI 2023 · Table 1
Domain adaptation and generalization
294.65protoLP*
ResNet-12
Transductive Few-Shot Learning With Prototype-Based Label Propagation by Iterative Graph Refinement
CVPR 2023 · CUB results table
Zero-/few-shot and meta-learning
394.02STL DeepBDC
ResNet-18
Joint Distribution Matters: Deep Brownian Distance Covariance for Few-Shot Classification
CVPR 2022 · Table: Results on fine-grained
Zero-/few-shot and meta-learning
490.77HFA (Ours)
ResNet18
Hyperbolic Feature Augmentation via Distribution Estimation and Infinite Sampling on Manifolds
NeurIPS 2022 · Table 1
Recognition and long-tail
589.5StarNet
2× ResNet-12
StarNet: towards Weakly Supervised Few-Shot Object Detection
AAAI 2021 · Few-shot classification table
Weakly supervised detection and segmentation
688.68LaplacianShot
ResNet-18
Laplacian Regularized Few-Shot Learning
ICML 2020 · Table 2
Zero-/few-shot and meta-learning
788.32P-Transfer (ProtoNet)
ResNet-12
Partial Is Better Than All: Revisiting Fine-tuning Strategy for Few-shot Learning
AAAI 2021 · Table 2
Zero-/few-shot and meta-learning
886.6DUAL ATT-NET
CNN-4
Dual Attention Networks for Few-Shot Fine-Grained Recognition
AAAI 2022 · Table 2
Zero-/few-shot and meta-learning
985.3COMETConcept Learners for Few-Shot Learning
ICLR 2021 · Table 1
Zero-/few-shot and meta-learning
1079.1CDKT + Cosine (ML) (tau<1)
Conv4
Revisiting Logistic-softmax Likelihood in Bayesian Meta-Learning for Few-Shot Classification
NeurIPS 2023 · Table 1
Uncertainty and Gaussian processes
7 more papers on this leaderboard, the baselines reported on it and a CSV download are free with a scime account.
Sign in free to see all

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