• Corpus ID: 215786215

Multi-Object Tracking with Siamese Track-RCNN

  title={Multi-Object Tracking with Siamese Track-RCNN},
  author={Bing Shuai and Andrew G. Berneshawi and Davide Modolo and Joseph Tighe},
Multi-object tracking systems often consist of a combination of a detector, a short term linker, a re-identification feature extractor and a solver that takes the output from these separate components and makes a final prediction. Differently, this work aims to unify all these in a single tracking system. Towards this, we propose Siamese Track-RCNN, a two stage detect-and-track framework which consists of three functional branches: (1) the detection branch localizes object instances; (2) the… 

Application of Multi-Object Tracking with Siamese Track-RCNN to the Human in Events Dataset

This work proposes Siamese Track-RCNN, a two stage detect-and-track framework which consists of three functional branches: (1) the detection branch localizes object instances; (2) theSiamese-based track branch estimates the object motion and (3) the object re-identification branch re-activates the previously terminated tracks when they re-emerge.

Multi-object Tracking with a Hierarchical Single-branch Network

An online multi-object tracking framework based on a hierarchical single-branch network that utilizes an improved Hierarchical Online Instance Matching (iHOIM) loss to explicitly model the interrelationship between object detection and Re-ID is proposed.

Similarity based person re-identification for multi-object tracking using deep Siamese network

A similarity-based person re-id framework, called SAT, is proposed, using a Siamese neural network via shared weights to improve the current state-of-the-art according to commonly used performance metrics with higher accuracy, less ID switches, less false positive and negative rates.

Compensation Tracker: Reprocessing Lost Object for Multi-Object Tracking

This paper analyzes the phenomenon of the lost tracking object in real-time tracking model on MOT2020 dataset and proposes a compensation tracker that can re-track missing tracking objects from lost objects and does not require additional networks so as to maintain speed-accuracy trade-off of the real- time model.

Siamese Visual Object Tracking: A Survey

An in-depth analysis of the core principles on which Siamese trackers operate with a discussion of incentives behind them is performed and an up-to-date qualitative and quantitative comparison of the prominent SiamesE trackers on established benchmarks is provided.

MOTR: End-to-End Multiple-Object Tracking with TRansformer

MOTR is proposed, which extends DETR and introduces “track query” to model the tracked instances in the entire video to enhance temporal relation modeling and serve as a stronger baseline for future research on temporal modeling and Transformer-based trackers.

MOTRv2: Bootstrapping End-to-End Multi-Object Tracking by Pretrained Object Detectors

This paper proposes MOTRv2, a simple yet effec-tive pipeline to bootstrap end-to-end multi-object tracking with a pretrained object detector, and aims to improve MOTR by elegantly incorporating an extra object detector.

SynDHN: Multi-Object Fish Tracker Trained on Synthetic Underwater Videos

  • M. A. MartijaP. Naval
  • Computer Science
    2020 25th International Conference on Pattern Recognition (ICPR)
  • 2021
This paper uses the Deep Hungarian Network (DHN) to repurpose DHN to become the tracking component of the algorithm by performing the task of affinity estimation between detector predictions, and considers both spatial and appearance features for affinity estimation.

Compensation Tracker: Data Association Method for Lost Object

Experiments show that after using the compensation tracker designed in this paper, evaluation indicators have improved in varying degrees on MOT Challenge data sets and the proposed method can effectively improve the tracking performance of the model.

Merging Tasks for Video Panoptic Segmentation

The task of video panoptic segmentation is studied and two different methods to solve the task will be proposed, which do not require training on a tailored VPS dataset to solve it.



Tracking Without Bells and Whistles

Overall, Tracktor yields superior tracking performance than any current tracking method and the analysis exposes remaining and unsolved tracking challenges to inspire future research directions.

Fusion of Head and Full-Body Detectors for Multi-object Tracking

This work demonstrates how to fuse two detectors into a tracking system using the Frank-Wolfe algorithm, and proposes to formulate tracking as a weighted graph labeling problem, resulting in a binary quadratic program.

High Performance Visual Tracking with Siamese Region Proposal Network

The Siamese region proposal network (Siamese-RPN) is proposed which is end-to-end trained off-line with large-scale image pairs for visual object tracking and consists of SiAMESe subnetwork for feature extraction and region proposal subnetwork including the classification branch and regression branch.

A Twofold Siamese Network for Real-Time Object Tracking

The proposed SA-Siam outperforms all other real-time trackers by a large margin on OTB-2013/50/100 benchmarks and proposes a channel attention mechanism for the semantic branch.

Fully-Convolutional Siamese Networks for Object Tracking

A basic tracking algorithm is equipped with a novel fully-convolutional Siamese network trained end-to-end on the ILSVRC15 dataset for object detection in video and achieves state-of-the-art performance in multiple benchmarks.

Exploit the Connectivity: Multi-Object Tracking with TrackletNet

This paper proposes an innovative and effective tracking method called TrackletNet Tracker (TNT) that combines temporal and appearance information together as a unified framework and achieves promising results on MOT16 and MOT17 benchmark datasets compared with other state-of-the-art methods.

MOTS: Multi-Object Tracking and Segmentation

This paper creates dense pixel-level annotations for two existing tracking datasets using a semi-automatic annotation procedure, and proposes a new baseline method which jointly addresses detection, tracking, and segmentation with a single convolutional network.

Fast Online Object Tracking and Segmentation: A Unifying Approach

This method improves the offline training procedure of popular fully-convolutional Siamese approaches for object tracking by augmenting their loss with a binary segmentation task, and operates online, producing class-agnostic object segmentation masks and rotated bounding boxes at 55 frames per second.

Motion Segmentation & Multiple Object Tracking by Correlation Co-Clustering

This work states this joint problem as a co-clustering problem that is principled and tractable by existing algorithms, and demonstrates the effectiveness of this approach by combining bottom-up motion segmentation by grouping of point trajectories with high-level multiple object tracking by clustering of bounding boxes.

Multiple Hypothesis Tracking Revisited

It is demonstrated that a classical MHT implementation from the 90's can come surprisingly close to the performance of state-of-the-art methods on standard benchmark datasets, and it is shown that appearance models can be learned efficiently via a regularized least squares framework.