Learning to Predict Visual Attributes in the Wild

TL;DR

This paper introduces a large-scale in-the-wild visual attribute dataset and proposes advanced techniques to improve multi-label attribute prediction, significantly outperforming existing methods.

Contribution

The paper presents a new extensive dataset for visual attribute prediction and novel methods addressing challenges like label sparsity and data imbalance.

Findings

Achieved 3.7 mAP improvement over state-of-the-art methods.

Developed a multi-hop attention-based CNN model.

Introduced a supervised attribute-aware contrastive learning algorithm.

Abstract

Visual attributes constitute a large portion of information contained in a scene. Objects can be described using a wide variety of attributes which portray their visual appearance (color, texture), geometry (shape, size, posture), and other intrinsic properties (state, action). Existing work is mostly limited to study of attribute prediction in specific domains. In this paper, we introduce a large-scale in-the-wild visual attribute prediction dataset consisting of over 927K attribute annotations for over 260K object instances. Formally, object attribute prediction is a multi-label classification problem where all attributes that apply to an object must be predicted. Our dataset poses significant challenges to existing methods due to large number of attributes, label sparsity, data imbalance, and object occlusion. To this end, we propose several techniques that systematically tackle these challenges, including a base model that utilizes both low- and high-level CNN features with multi-hop attention, reweighting and resampling techniques, a novel negative label expansion scheme, and a novel supervised attribute-aware contrastive learning algorithm. Using these techniques, we achieve near 3.7 mAP and 5.7 overall F1 points improvement over the current state of the art. Further details about the VAW dataset can be found at http://vawdataset.com/.