Bird Species Categorization Using Pose Normalized Deep Convolutional Nets

TL;DR

This paper introduces a pose-normalized deep learning architecture for bird species classification that achieves near expert-level accuracy by combining pose estimation, local feature extraction, and deep convolutional networks.

Contribution

It presents a novel pose normalization scheme using graph-based clustering and integrates multi-level features for improved fine-grained bird classification.

Findings

Achieved 75% accuracy in bird species recognition, surpassing previous methods (55-65%).

Demonstrated effectiveness of pose normalization and multi-level feature integration.

Provided empirical analysis of pose normalization schemes and deep feature extraction techniques.

Abstract

We propose an architecture for fine-grained visual categorization that approaches expert human performance in the classification of bird species. Our architecture first computes an estimate of the object's pose; this is used to compute local image features which are, in turn, used for classification. The features are computed by applying deep convolutional nets to image patches that are located and normalized by the pose. We perform an empirical study of a number of pose normalization schemes, including an investigation of higher order geometric warping functions. We propose a novel graph-based clustering algorithm for learning a compact pose normalization space. We perform a detailed investigation of state-of-the-art deep convolutional feature implementations and fine-tuning feature learning for fine-grained classification. We observe that a model that integrates lower-level feature layers with pose-normalized extraction routines and higher-level feature layers with unaligned image features works best. Our experiments advance state-of-the-art performance on bird species recognition, with a large improvement of correct classification rates over previous methods (75% vs. 55-65%).