How Far are We from Solving Pedestrian Detection?

Abstract

Encouraged by the recent progress in pedestrian detection, we investigate the gap between current state-of-the-art methods and the "perfect single frame detector". We enable our analysis by creating a human baseline for pedestrian detection (over the Caltech dataset), and by manually clustering the recurrent errors of a top detector. Our results characterize both localization and background-versus-foreground errors. To address localization errors we study the impact of training annotation noise on the detector performance, and show that we can improve even with a small portion of sanitized training data. To address background/foreground discrimination, we study convnets for pedestrian detection, and discuss which factors affect their performance. Other than our in-depth analysis, we report top performance on the Caltech dataset, and provide a new sanitized set of training and test annotations.

CVPR16 camera ready

References in corpus (7)

• Ten Years of Pedestrian Detection, What Have We Learned?
• Analyzing the Performance of Multilayer Neural Networks for Object Recognition
• Local Decorrelation For Improved Detection
• Filtered Channel Features for Pedestrian Detection
• Taking a Deeper Look at Pedestrians
• Exploring Human Vision Driven Features for Pedestrian Detection
• Pedestrian Detection aided by Deep Learning Semantic Tasks

Cited by in corpus (28)

• Person Re-identification: Past, Present and Future
• Too Far to See? Not Really! --- Pedestrian Detection with Scale-aware Localization Policy
• Perceptual Generative Adversarial Networks for Small Object Detection
• Illuminating Pedestrians via Simultaneous Detection & Segmentation
• Pedestrian-Synthesis-GAN: Generating Pedestrian Data in Real Scene and Beyond
• CityPersons: A Diverse Dataset for Pedestrian Detection
• Multispectral Deep Neural Networks for Pedestrian Detection
• NoScope: Optimizing Neural Network Queries over Video at Scale
• Person Re-identification in the Wild
• Person Search via A Mask-Guided Two-Stream CNN Model
• Repulsion Loss: Detecting Pedestrians in a Crowd
• ViTAA: Visual-Textual Attributes Alignment in Person Search by Natural Language
• Where, What, Whether: Multi-modal Learning Meets Pedestrian Detection
• PCN: Part and Context Information for Pedestrian Detection with CNNs
• Double Anchor R-CNN for Human Detection in a Crowd
• A Survey on Deep Domain Adaptation and Tiny Object Detection Challenges, Techniques and Datasets
• Unsupervised Hard Example Mining from Videos for Improved Object Detection
• PedHunter: Occlusion Robust Pedestrian Detector in Crowded Scenes
• Is Faster R-CNN Doing Well for Pedestrian Detection?
• Vulnerable road user detection: state-of-the-art and open challenges
• Exploring Multi-Branch and High-Level Semantic Networks for Improving Pedestrian Detection
• Relational Learning for Joint Head and Human Detection
• Pedestrian Detection with Autoregressive Network Phases
• What Can Help Pedestrian Detection?
• Mutual-Supervised Feature Modulation Network for Occluded Pedestrian Detection
• Robustness Analysis of Pedestrian Detectors for Surveillance
• Part-Level Convolutional Neural Networks for Pedestrian Detection Using Saliency and Boundary Box Alignment
• Efficient Learning of Pinball TWSVM using Privileged Information and its applications