Distractor-Aware Fast Tracking via Dynamic Convolutions and MOT Philosophy

  title={Distractor-Aware Fast Tracking via Dynamic Convolutions and MOT Philosophy},
  author={Zikai Zhang and Bineng Zhong and Shengping Zhang and Zhenjun Tang and Xin Liu and Zhaoxiang Zhang},
  journal={2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
  • Zikai Zhang, Bineng Zhong, +3 authors Zhaoxiang Zhang
  • Published 25 April 2021
  • Computer Science
  • 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
A practical long-term tracker typically contains three key properties, i.e. an efficient model design, an effective global re-detection strategy and a robust distractor awareness mechanism. However, most state-of-the-art long-term trackers (e.g., Pseudo and re-detecting based ones) do not take all three key properties into account and therefore may either be time-consuming or drift to distractors. To address the issues, we propose a two-task tracking framework (named DMTrack), which utilizes… 

Figures and Tables from this paper

Learning Target Candidate Association to Keep Track of What Not to Track
This work proposes a training strategy that combines partial annotations with self-supervision to keep track of distractor objects in order to continue tracking the target, and introduces a learned association network to propagate the identities of all target candidates from frameto-frame.


Distractor-aware Siamese Networks for Visual Object Tracking
This paper focuses on learning distractor-aware Siamese networks for accurate and long-term tracking, and extends the proposed approach for long- term tracking by introducing a simple yet effective local-to-global search region strategy.
Bridging the Gap Between Detection and Tracking: A Unified Approach
This paper aims to explore a general framework for building trackers directly upon almost any advanced object detector, and introduces an anchored updating strategy to alleviate the problem of overfitting.
High-Performance Long-Term Tracking With Meta-Updater
This work proposes a novel offline-trained Meta-Updater that can effectively integrate geometric, discriminative, and appearance cues in a sequential manner, and then mine the sequential information with a designed cascaded LSTM module.
GlobalTrack: A Simple and Strong Baseline for Long-term Tracking
This work proposes GlobalTrack, a pure global instance search based tracker that makes no assumption on the temporal consistency of the target's positions and scales, and it is able to perform full-image and multi-scale search of arbitrary instances with only a single query as the guide.
Performance Evaluation Methodology for Long-Term Single-Object Tracking
A long-term visual object tracking performance evaluation methodology and a benchmark are proposed and it is shown that these measures generalize the short-term performance measures, thus linking the two tracking problems.
‘Skimming-Perusal’ Tracking: A Framework for Real-Time and Robust Long-Term Tracking
This work presents a novel robust and real-time long-term tracking framework based on the proposed skimming and perusal modules, designed to efficiently choose the most possible regions from a large number of sliding windows for image-wide global search.
Siam R-CNN: Visual Tracking by Re-Detection
This work presents Siam R-CNN, a Siamese re-detection architecture which unleashes the full power of two-stage object detection approaches for visual object tracking, and combines this with a novel tracklet-based dynamic programming algorithm to model the full history of both the object to be tracked and potential distractor objects.
Hierarchical Tracking by Reinforcement Learning-Based Searching and Coarse-to-Fine Verifying
This work proposes a hierarchical tracker that learns to move and track based on the combination of data-driven search at the coarse level and coarse-to-fine verification at the fine level, and utilizes a recurrent convolutional neural network-based deep Q-network to effectively learn data- driven searching policies.
Learning regression and verification networks for long-term visual tracking
This work proposes a novel long-term tracking framework based on deep regression and verification networks, designed using the object-aware feature fusion and region proposal networks to generate a series of candidates and estimate their similarity scores effectively.
Siamese Box Adaptive Network for Visual Tracking
SiamBAN views the visual tracking problem as a parallel classification and regression problem, and thus directly classifies objects and regresses their bounding boxes in a unified FCN, making SiamB Ban more flexible and general.