Fine-grained Apparel Classification and Retrieval without rich annotations

TL;DR

This paper introduces a robust apparel classification and retrieval framework that operates without requiring detailed annotations like bounding boxes or landmarks, using compact bilinear CNNs and triplet loss for improved accuracy across domains.

Contribution

The proposed method eliminates the need for rich annotations and bounding box detectors, enabling direct training on category labels and triplets for apparel classification and retrieval.

Findings

Achieves competitive accuracy on DeepFashion datasets.

Effective in cross-domain retrieval with varied image conditions.

Reduces annotation dependency compared to previous methods.

Abstract

The ability to correctly classify and retrieve apparel images has a variety of applications important to e-commerce, online advertising and internet search. In this work, we propose a robust framework for fine-grained apparel classification, in-shop and cross-domain retrieval which eliminates the requirement of rich annotations like bounding boxes and human-joints or clothing landmarks, and training of bounding box/ key-landmark detector for the same. Factors such as subtle appearance differences, variations in human poses, different shooting angles, apparel deformations, and self-occlusion add to the challenges in classification and retrieval of apparel items. Cross-domain retrieval is even harder due to the presence of large variation between online shopping images, usually taken in ideal lighting, pose, positive angle and clean background as compared with street photos captured by users in complicated conditions with poor lighting and cluttered scenes. Our framework uses compact bilinear CNN with tensor sketch algorithm to generate embeddings that capture local pairwise feature interactions in a translationally invariant manner. For apparel classification, we pass the feature embeddings through a softmax classifier, while, the in-shop and cross-domain retrieval pipelines use a triplet-loss based optimization approach, such that squared Euclidean distance between embeddings measures the dissimilarity between the images. Unlike previous works that relied on bounding box, key clothing landmarks or human joint detectors to assist the final deep classifier, proposed framework can be trained directly on the provided category labels or generated triplets for triplet loss optimization. Lastly, Experimental results on the DeepFashion fine-grained categorization, and in-shop and consumer-to-shop retrieval datasets provide a comparative analysis with previous work performed in the domain.