Pedestrian Detection: Domain Generalization, CNNs, Transformers and Beyond

01/10/2022
by   Irtiza Hasan, et al.
8

Pedestrian detection is the cornerstone of many vision based applications, starting from object tracking to video surveillance and more recently, autonomous driving. With the rapid development of deep learning in object detection, pedestrian detection has achieved very good performance in traditional single-dataset training and evaluation setting. However, in this study on generalizable pedestrian detectors, we show that, current pedestrian detectors poorly handle even small domain shifts in cross-dataset evaluation. We attribute the limited generalization to two main factors, the method and the current sources of data. Regarding the method, we illustrate that biasness present in the design choices (e.g anchor settings) of current pedestrian detectors are the main contributing factor to the limited generalization. Most modern pedestrian detectors are tailored towards target dataset, where they do achieve high performance in traditional single training and testing pipeline, but suffer a degrade in performance when evaluated through cross-dataset evaluation. Consequently, a general object detector performs better in cross-dataset evaluation compared with state of the art pedestrian detectors, due to its generic design. As for the data, we show that the autonomous driving benchmarks are monotonous in nature, that is, they are not diverse in scenarios and dense in pedestrians. Therefore, benchmarks curated by crawling the web (which contain diverse and dense scenarios), are an efficient source of pre-training for providing a more robust representation. Accordingly, we propose a progressive fine-tuning strategy which improves generalization. Code and models cab accessed at https://github.com/hasanirtiza/Pedestron.

READ FULL TEXT

page 5

page 9

page 10

research
03/19/2020

Pedestrian Detection: The Elephant In The Room

Pedestrian detection is used in many vision based applications ranging f...
research
01/23/2015

Taking a Deeper Look at Pedestrians

In this paper we study the use of convolutional neural networks (convnet...
research
04/05/2019

High-level Semantic Feature Detection:A New Perspective for Pedestrian Detection

Object detection generally requires sliding-window classifiers in tradit...
research
12/12/2022

Comparison Of Deep Object Detectors On A New Vulnerable Pedestrian Dataset

Pedestrian safety is one primary concern in autonomous driving. The unde...
research
08/19/2020

Anchor-free Small-scale Multispectral Pedestrian Detection

Multispectral images consisting of aligned visual-optical (VIS) and ther...
research
07/12/2018

Robustness Analysis of Pedestrian Detectors for Surveillance

To obtain effective pedestrian detection results in surveillance video, ...
research
04/26/2021

Variational Pedestrian Detection

Pedestrian detection in a crowd is a challenging task due to a high numb...

Please sign up or login with your details

Forgot password? Click here to reset