Mooyeol Baek

TL;DR: The Visual Object Tracking challenge VOT2016 goes beyond its predecessors by introducing a new semi-automatic ground truth bounding box annotation methodology and extending the evaluation system with the no-reset experiment.

TL;DR: An online visual tracking algorithm by managing multiple target appearance models in a tree structure using Convolutional Neural Networks to represent target appearances, where multiple CNNs collaborate to estimate target states and determine the desirable paths for online model updates in the tree.

TL;DR: This work proposes Quadruplet Convolutional Neural Networks (Quad-CNN) for multi-object tracking, which learn to associate object detections across frames using quadruplet losses and employs a multi-task loss to jointly learn object association and bounding box regression for better localization.

TL;DR: This work presents a fast and accurate visual tracking algorithm based on the multi-domain convolutional neural network (MDNet) that accelerates feature extraction procedure and learns more discriminative models for instance classification; it enhances representation quality of target and background by maintaining a high resolution feature map with a large receptive field per activation.

TL;DR: In this paper, a multi-domain convolutional neural network (MDNet) is proposed to accelerate feature extraction procedure and learn more discriminative models for instance classification; it enhances representation quality of target and background by maintaining a high resolution feature map with a large receptive field per activation.