Abstract: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 can accessed at <a class="link-external link-https" href="https://github.com/hasanirtiza/Pedestron" rel="external noopener nofollow">this https URL</a>.

Transfer Learning Based on Sparse Coding and It's Application in Pedestrian Detection

A Pedestrian Classification Method Based on Transfer Learning

Pedestrian Detection Based on Sparse Coding and Transfer Learning

A Novel Approach to Design the Fast Pedestrian Detection for Video Surveillance System

Transfer learning for pedestrian detection

Pedestrian Detection by Using CNN Features with Skip Connection.

Adaptive Pedestrian Detection by Predicting Classifier

Cognitive Pedestrian Detector: Adapting Detector to Specific Scene by Transferring Attributes.

Pedestrian Detection aided by Deep Learning Semantic Tasks

Towards Accurate Dense Pedestrian Detection Via Occlusion-Prediction Aware Label Assignment and Hierarchical-Nms.

Deep Learning-Based Pedestrian Detection Using RGB Images and Sparse LiDAR Point Clouds

Discriminative Latent Semantic Feature Learning for Pedestrian Detection

A Study on Pedestrian Re-identification Based on Transfer Learning

Feature Selection And Pedestrian Detection Based On Sparse Representation

Pedestrian Detection: Domain Generalization, CNNs, Transformers and Beyond

Deep Convolutional Neural Networks For Pedestrian Detection With Skip Pooling

Adaptive pedestrian detection using convolutional neural network with dynamically adjusted classifier.

Multi-pose Pedestrian Detection Based on Posterior Multiple Sparse Dictionaries

R-SSD: Refined Single Shot Multibox Detector for Pedestrian Detection

An Improved Real-time Detection and Localization Scheme for Pedestrian Based on Information Fusion

AutoPedestrian: An Automatic Data Augmentation and Loss Function Search Scheme for Pedestrian Detection