# Filtered Channel Features for Pedestrian Detection

arXiv: https://arxiv.org/abs/1501.05759 · HTML: https://mlsanity.com/paper/1501.05759/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 23 Jan 2015.

## Built on by (most cited first, 11 of 11)

- [Weakly Aligned Feature Fusion for Multimodal Object Detection](https://mlsanity.com/paper/2204.09848/) (2022)
- [NMS-Loss: Learning with Non-Maximum Suppression for Crowded Pedestrian Detection](https://mlsanity.com/paper/2106.02426/) (2021)
- [Vulnerable Road User Detection and Safety Enhancement: A Comprehensive Survey](https://mlsanity.com/paper/2405.19202/) (2024)
- [Pedestrian Detection by Exemplar-Guided Contrastive Learning](https://mlsanity.com/paper/2111.08974/) (2021)
- [Nighttime Pedestrian Detection Based on Fore-Background Contrast Learning](https://mlsanity.com/paper/2408.03030/) (2024)
- [Deep Learning Techniques for Visual Counting](https://mlsanity.com/paper/2206.03033/) (2022)
- [Improving Urban Flood Prediction using LSTM-DeepLabv3+ and Bayesian Optimization with Spatiotemporal feature fusion](https://mlsanity.com/paper/2304.09994/) (2023)
- [Towards Driver Behavior Understanding: Weakly-Supervised Risk Perception in Driving Scenes](https://mlsanity.com/paper/2603.05926/) (2026)
- [Efficient Learning of Pinball TWSVM using Privileged Information and its applications](https://mlsanity.com/paper/2107.06744/) (2021)
- [Application-Driven AI Paradigm for Person Counting in Various Scenarios](https://mlsanity.com/paper/2303.13788/) (2023)
- [BAdaCost: Multi-class Boosting with Costs](https://mlsanity.com/paper/2402.04465/) (2024)

## Abstract (arXiv)

This paper starts from the observation that multiple top performing pedestrian detectors can be modelled by using an intermediate layer filtering low-level features in combination with a boosted decision forest. Based on this observation we propose a unifying framework and experimentally explore different filter families. We report extensive results enabling a systematic analysis. Using filtered channel features we obtain top performance on the challenging Caltech and KITTI datasets, while using only HOG+LUV as low-level features. When adding optical flow features we further improve detection quality and report the best known results on the Caltech dataset, reaching 93% recall at 1 FPPI.