VITAL: VIsual Tracking via Adversarial Learning

Abstract

The tracking-by-detection framework consists of two stages, i.e., drawing samples around the target object in the first stage and classifying each sample as the target object or as background in the second stage. The performance of existing trackers using deep classification networks is limited by two aspects. First, the positive samples in each frame are highly spatially overlapped, and they fail to capture rich appearance variations. Second, there exists extreme class imbalance between positive and negative samples. This paper presents the VITAL algorithm to address these two problems via adversarial learning. To augment positive samples, we use a generative network to randomly generate masks, which are applied to adaptively dropout input features to capture a variety of appearance changes. With the use of adversarial learning, our network identifies the mask that maintains the most robust features of the target objects over a long temporal span. In addition, to handle the issue of class imbalance, we propose a high-order cost sensitive loss to decrease the effect of easy negative samples to facilitate training the classification network. Extensive experiments on benchmark datasets demonstrate that the proposed tracker performs favorably against state-of-the-art approaches.

arXiv commentSpotlight in CVPR 2018

References in corpus (10)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Focal Loss for Dense Object Detection
• A Novel Performance Evaluation Methodology for Single-Target Trackers
• f-GAN: Training Generative Neural Samplers using Variational Divergence Minimization
• Siamese Instance Search for Tracking
• CREST: Convolutional Residual Learning for Visual Tracking
• Stylizing Face Images via Multiple Exemplars
• Learning to Hallucinate Face Images via Component Generation and Enhancement
• DeLiGAN : Generative Adversarial Networks for Diverse and Limited Data
• Fast Preprocessing for Robust Face Sketch Synthesis

Cited by in corpus (11)

• A Review on Generative Adversarial Networks: Algorithms, Theory, and Applications
• Deeper and Wider Siamese Networks for Real-Time Visual Tracking
• Unsupervised Deep Tracking
• Siamese Cascaded Region Proposal Networks for Real-Time Visual Tracking
• Joint Group Feature Selection and Discriminative Filter Learning for Robust Visual Object Tracking
• GlobalTrack: A Simple and Strong Baseline for Long-term Tracking
• A Review of Visual Trackers and Analysis of its Application to Mobile Robot
• Deep Attentive Tracking via Reciprocative Learning
• Improving Object Detection with Inverted Attention
• IoU Attack: Towards Temporally Coherent Black-Box Adversarial Attack for Visual Object Tracking
• ROAM: Recurrently Optimizing Tracking Model