DeepAI AI Chat
Log In Sign Up

Revisiting Pose-Normalization for Fine-Grained Few-Shot Recognition

by   Luming Tang, et al.
cornell university

Few-shot, fine-grained classification requires a model to learn subtle, fine-grained distinctions between different classes (e.g., birds) based on a few images alone. This requires a remarkable degree of invariance to pose, articulation and background. A solution is to use pose-normalized representations: first localize semantic parts in each image, and then describe images by characterizing the appearance of each part. While such representations are out of favor for fully supervised classification, we show that they are extremely effective for few-shot fine-grained classification. With a minimal increase in model capacity, pose normalization improves accuracy between 10 and 20 percentage points for shallow and deep architectures, generalizes better to new domains, and is effective for multiple few-shot algorithms and network backbones. Code is available at


page 2

page 4

page 8

page 13


Fine-Grained Object Classification via Self-Supervised Pose Alignment

Semantic patterns of fine-grained objects are determined by subtle appea...

ePillID Dataset: A Low-Shot Fine-Grained Benchmark for Pill Identification

Identifying prescription medications is a frequent task for patients and...

Bird Species Categorization Using Pose Normalized Deep Convolutional Nets

We propose an architecture for fine-grained visual categorization that a...

SnapMix: Semantically Proportional Mixing for Augmenting Fine-grained Data

Data mixing augmentation has proved effective in training deep models. R...

Data-Centric AI Paradigm Based on Application-Driven Fine-grained Dataset Design

Deep learning has a wide range of applications in industrial scenario, b...

VPN++: Rethinking Video-Pose embeddings for understanding Activities of Daily Living

Many attempts have been made towards combining RGB and 3D poses for the ...

Focus Longer to See Better:Recursively Refined Attention for Fine-Grained Image Classification

Deep Neural Network has shown great strides in the coarse-grained image ...

Code Repositories


(CVPR 2020) Revisiting Pose-Normalization for Fine-Grained Few-Shot Recognition

view repo