• Home
  • Search
  • Occluded Pedestrian Detection Through Guided Attention in CNNs
  • Cite Icon466
  • https://doi.org/10.1109/cvpr.2018.00731Copy DOI Icon

Occluded Pedestrian Detection Through Guided Attention in CNNs

  • Jun 1, 2018
  • Shanshan Zhang +2 more
Show More
  • Abstract
  • Literature Map
  • References
  • Citations
  • Similar Papers
Abstract

Pedestrian detection has progressed significantly in the last years. However, occluded people are notoriously hard to detect, as their appearance varies substantially depending on a wide range of occlusion patterns. In this paper, we aim to propose a simple and compact method based on the FasterRCNN architecture for occluded pedestrian detection. We start with interpreting CNN channel features of a pedestrian detector, and we find that different channels activate responses for different body parts respectively. These findings motivate us to employ an attention mechanism across channels to represent various occlusion patterns in one single model, as each occlusion pattern can be formulated as some specific combination of body parts. Therefore, an attention network with self or external guidances is proposed as an add-on to the baseline FasterRCNN detector. When evaluating on the heavy occlusion subset, we achieve a significant improvement of 8pp to the baseline FasterRCNN detector on CityPersons and on Caltech we outperform the state-of-the-art method by 4pp.

Similar Papers
  • Research Article
  • Citations3

Replacing the human driver: An objective benchmark for occluded pedestrian detection

  • Jul 18, 2023
  • Biomimetic Intelligence and Robotics
  • Shane Gilroy +4
  • Research Article
  • Citations32

PRNet++: Learning towards generalized occluded pedestrian detection via progressive refinement network

  • Jan 21, 2022
  • Neurocomputing
  • Xiaolin Song +4
  • PDF
  • Research Article
  • Citations15

Joint Pedestrian and Body Part Detection via Semantic Relationship Learning

  • Feb 21, 2019
  • Applied Sciences
  • Junhua Gu +3
  • Conference Article
  • Citations6

Mutual-Supervised Feature Modulation Network for Occluded Pedestrian Detection

  • Jan 10, 2021
  • Ye He +2
  • Research Article
  • Citations13

A Unified Multi-Task Learning Architecture for Fast and Accurate Pedestrian Detection

  • Sep 04, 2020
  • IEEE Transactions on Intelligent Transportation Systems
  • Chengju Zhou +2
  • Research Article
  • Citations61

Pedestrian Detection via Body Part Semantic and Contextual Information With DNN

  • Nov 01, 2018
  • IEEE Transactions on Multimedia
  • Shiguang Wang +4
  • Conference Article
  • Citations81

Semantic Channels for Fast Pedestrian Detection

  • Jun 01, 2016
  • Arthur Daniel Costea +1
  • Conference Article
  • Citations5

Pedestrian Detection with Local Feature Assistant

  • May 01, 2007
  • Y.W Xu +2
  • Research Article

Attention-based digital filter with anchor-free feature pyramid learning model for pedestrian detection

  • Apr 18, 2024
  • Journal of Intelligent & Fuzzy Systems
  • Ankit Shrivastava +1
  • Conference Article
  • Citations74

Parsing Occluded People

  • Jun 01, 2014
  • Golnaz Ghiasi +3
  • Conference Article
  • Citations7

Part-level fully convolutional networks for pedestrian detection

  • Mar 01, 2017
  • Xinran Wang +2
  • PDF
  • Research Article
  • Citations1

A Layered Approach for Robust Spatial Virtual Human Pose Reconstruction Using a Still Image

  • Feb 20, 2016
  • Sensors (Basel, Switzerland)
  • Chengyu Guo +3
  • Research Article
  • Citations54

Semantic head enhanced pedestrian detection in a crowd

  • Mar 16, 2020
  • Neurocomputing
  • Ruiqi Lu +2
  • Research Article
  • Citations25

Efficient Single-Stage Pedestrian Detector by Asymptotic Localization Fitting and Multi-Scale Context Encoding.

  • Sep 16, 2019
  • IEEE Transactions on Image Processing
  • Wei Liu +2
  • Conference Article
  • Citations1

YOLO11-HP: a high-performance YOLOv11 for occluded pedestrian detection

  • Sep 08, 2025
  • Hengfeng Gong
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.