# Continuity-Discrimination Convolutional Neural Network for Visual Object Tracking

arXiv: https://arxiv.org/abs/2104.08739 · HTML: https://mlsanity.com/paper/2104.08739/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 8 of its 28 references that are in the map.
- First posted on arXiv 18 Apr 2021.

## Builds on (most cited first, 8 of 8)

- [Fully-Convolutional Siamese Networks for Object Tracking](https://mlsanity.com/paper/1606.09549/) (2016)
- [Staple: Complementary Learners for Real-Time Tracking](https://mlsanity.com/paper/1512.01355/) (2015)
- [Siamese Instance Search for Tracking](https://mlsanity.com/paper/1605.05863/) (2016)
- [Discriminative Correlation Filter with Channel and Spatial Reliability](https://mlsanity.com/paper/1611.08461/) (2016)
- [Online Tracking by Learning Discriminative Saliency Map with Convolutional Neural Network](https://mlsanity.com/paper/1502.06796/) (2015)
- [A Survey of Appearance Models in Visual Object Tracking](https://mlsanity.com/paper/1303.4803/) (2013)
- [Transferring Rich Feature Hierarchies for Robust Visual Tracking](https://mlsanity.com/paper/1501.04587/) (2015)
- [PVANET: Deep but Lightweight Neural Networks for Real-time Object Detection](https://mlsanity.com/paper/1608.08021/) (2016)

## Abstract (arXiv)

This paper proposes a novel model, named Continuity-Discrimination Convolutional Neural Network (CD-CNN), for visual object tracking. Existing state-of-the-art tracking methods do not deal with temporal relationship in video sequences, which leads to imperfect feature representations. To address this problem, CD-CNN models temporal appearance continuity based on the idea of temporal slowness. Mathematically, we prove that, by introducing temporal appearance continuity into tracking, the upper bound of target appearance representation error can be sufficiently small with high probability. Further, in order to alleviate inaccurate target localization and drifting, we propose a novel notion, object-centroid, to characterize not only objectness but also the relative position of the target within a given patch. Both temporal appearance continuity and object-centroid are jointly learned during offline training and then transferred for online tracking. We evaluate our tracker through extensive experiments on two challenging benchmarks and show its competitive tracking performance compared with state-of-the-art trackers.