• Corpus ID: 212725600

Refinements in Motion and Appearance for Online Multi-Object Tracking

  title={Refinements in Motion and Appearance for Online Multi-Object Tracking},
  author={Piao Huang and Shoudong Han and Jun Zhao and Donghaisheng Liu and Hongwei Wang and En Yu and Alex Chichung Kot},
Modern multi-object tracking (MOT) system usually involves separated modules, such as motion model for location and appearance model for data association. However, the compatible problems within both motion and appearance models are always ignored. In this paper, a general architecture named as MIF is presented by seamlessly blending the Motion integration, three-dimensional(3D) Integral image and adaptive appearance feature Fusion. Since the uncertain pedestrian and camera motions are usually… 

Figures and Tables from this paper

Online multi-object tracking using multi-function integration and tracking simulation training

An end-to-end training method is proposed to simulate the multi-object tracking process during the training and expand the training data by using the historical position of the target combined with the prediction of the motion model.

Simple Unsupervised Multi-Object Tracking

This work proposes an unsupervised re-identification network, thus sidestepping the labeling costs entirely, required for training, and demonstrates that SimpleReID performs substantially better than simpler alternatives, and recovers the full performance of its supervised counterpart consistently across diverse tracking frameworks.

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.

Detecting Invisible People

This work re-purpose tracking benchmarks and proposes new metrics for the task of detecting invisible objects, focusing on the illustrative case of people, and is the first work to demonstrate the effectiveness of monocular depth estimation for thetask of tracking and detecting occluded objects.

Adaptive Tracking Algorithm for Multiperson Motion Targets Based on Local Feature Similarity

The experimental results show that combining the multiperson moving target adaptive tracking algorithm based on local feature similarity into the DSST model can improve the average accuracy and success rate of theDSST model and has good stability.

FTN: Foreground-Guided Texture-Focused Person Re-Identification

A Foreground-Guided Texture-Focused Network (FTN) for Re-ID, which can weaken the representation of unrelated background and highlight the attributes person-related in an end-to-end manner and performs favorably against the state-of-the-art methods.

Leveraging Structural Cues for Better Training and Deployment in Computer Vision

Surprisingly, it is found that the classical Conditional Risk Minimization (CRM) method is a far more suitable alternative than the works on deep hierarchy-aware classification; CRM preserves the base model’s top-1 accuracy and brings the most likely predictions of the model closer to the ground truth and is able to provide reliable probability estimates.

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.



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.

Multiple Target Tracking in World Coordinate with Single, Minimally Calibrated Camera

A novel algorithm for multi-object tracking that is capable of working under very challenging conditions such as minimal hardware equipment, uncalibrated monocular camera, occlusions and severe background clutter is proposed.

Aggregate Tracklet Appearance Features for Multi-Object Tracking

A multitask convolutional neural network is proposed to learn the discriminative features and spatial-temporal attentions jointly and decompose an object in a static image with spatial attentions, and then aggregate multiple features in a tracklet based on the temporal attentions.

Online Multi-Object Tracking With Instance-Aware Tracker and Dynamic Model Refreshment

This paper proposes an instance-aware tracker to integrate SOT techniques for MOT by encoding awareness both within and between target models, and considers response maps from all target models and assigns spatial locations exclusively to optimize the overall accuracy.

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.

Spatial-Temporal Relation Networks for Multi-Object Tracking

A unified framework for similarity measurement based on spatial-temporal relation network which could simultaneously encode various cues and perform reasoning across both spatial and temporal domains is presented.

Online Multi-Object Tracking with Dual Matching Attention Networks

This paper introduces a cost-sensitive tracking loss based on the state-of-the-art visual tracker which encourages the model to focus on hard negative distractors during online learning and proposes Dual Matching Attention Networks (DMAN) with both spatial and temporal attention mechanisms.

FAMNet: Joint Learning of Feature, Affinity and Multi-Dimensional Assignment for Online Multiple Object Tracking

  • Peng ChuHaibin Ling
  • Computer Science
    2019 IEEE/CVF International Conference on Computer Vision (ICCV)
  • 2019
An end-to-end model, named FAMNet, where Feature extraction, Affinity estimation and Multi-dimensional assignment are refined in a single network, where all layers in FAMNet are designed differentiable thus can be optimized jointly to learn the discriminative features and higher-order affinity model for robust MOT.

Tracklet Association Tracker: An End-to-End Learning-based Association Approach for Multi-Object Tracking

A new multiple object tracking framework with data-driven association method, named as Tracklet Association Tracker (TAT), which aims at gluing feature learning and data association into a unity by a bi-level optimization formulation so that the association results can be directly learned from features.

Temporal dynamic appearance modeling for online multi-person tracking