Learning by tracking: Siamese CNN for robust target association

04/26/2016
by   Laura Leal-Taixé, et al.
0

This paper introduces a novel approach to the task of data association within the context of pedestrian tracking, by introducing a two-stage learning scheme to match pairs of detections. First, a Siamese convolutional neural network (CNN) is trained to learn descriptors encoding local spatio-temporal structures between the two input image patches, aggregating pixel values and optical flow information. Second, a set of contextual features derived from the position and size of the compared input patches are combined with the CNN output by means of a gradient boosting classifier to generate the final matching probability. This learning approach is validated by using a linear programming based multi-person tracker showing that even a simple and efficient tracker may outperform much more complex models when fed with our learned matching probabilities. Results on publicly available sequences show that our method meets state-of-the-art standards in multiple people tracking.

READ FULL TEXT

page 1

page 3

research
10/30/2018

Multimodal matching using a Hybrid Convolutional Neural Network

In this work we propose a novel Convolutional Neural Network (CNN) archi...
research
08/22/2022

Minkowski Tracker: A Sparse Spatio-Temporal R-CNN for Joint Object Detection and Tracking

Recent research in multi-task learning reveals the benefit of solving re...
research
11/27/2017

Hierarchical Siamese Network for Thermal Infrared Object Tracking

Most thermal infrared (TIR) tracking methods are discriminative, which t...
research
04/02/2021

Learning to Filter: Siamese Relation Network for Robust Tracking

Despite the great success of Siamese-based trackers, their performance u...
research
01/24/2020

Learning a distance function with a Siamese network to localize anomalies in videos

This work introduces a new approach to localize anomalies in surveillanc...
research
12/06/2015

PatchBatch: a Batch Augmented Loss for Optical Flow

We propose a new pipeline for optical flow computation, based on Deep Le...
research
08/02/2021

Forward-Looking Sonar Patch Matching: Modern CNNs, Ensembling, and Uncertainty

Application of underwater robots are on the rise, most of them are depen...

Please sign up or login with your details

Forgot password? Click here to reset