Learning Complexity-Aware Cascades for Deep Pedestrian Detection

Abstract

The design of complexity-aware cascaded detectors, combining features of very different complexities, is considered. A new cascade design procedure is introduced, by formulating cascade learning as the Lagrangian optimization of a risk that accounts for both accuracy and complexity. A boosting algorithm, denoted as complexity aware cascade training (CompACT), is then derived to solve this optimization. CompACT cascades are shown to seek an optimal trade-off between accuracy and complexity by pushing features of higher complexity to the later cascade stages, where only a few difficult candidate patches remain to be classified. This enables the use of features of vastly different complexities in a single detector. In result, the feature pool can be expanded to features previously impractical for cascade design, such as the responses of a deep convolutional neural network (CNN). This is demonstrated through the design of a pedestrian detector with a pool of features whose complexities span orders of magnitude. The resulting cascade generalizes the combination of a CNN with an object proposal mechanism: rather than a pre-processing stage, CompACT cascades seamlessly integrate CNNs in their stages. This enables state of the art performance on the Caltech and KITTI datasets, at fairly fast speeds.

References in corpus (5)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Caffe: Convolutional Architecture for Fast Feature Embedding
• Ten Years of Pedestrian Detection, What Have We Learned?
• Filtered Channel Features for Pedestrian Detection
• Taking a Deeper Look at Pedestrians

Cited by in corpus (27)

• Focus: Querying Large Video Datasets with Low Latency and Low Cost
• Too Far to See? Not Really! --- Pedestrian Detection with Scale-aware Localization Policy
• Perceptual Generative Adversarial Networks for Small Object Detection
• Object Detection with Deep Learning: A Review
• Illuminating Pedestrians via Simultaneous Detection & Segmentation
• CityPersons: A Diverse Dataset for Pedestrian Detection
• Deciding How to Decide: Dynamic Routing in Artificial Neural Networks
• NoScope: Optimizing Neural Network Queries over Video at Scale
• Person Re-identification in the Wild
• Occlusion-aware R-CNN: Detecting Pedestrians in a Crowd
• IDK Cascades: Fast Deep Learning by Learning not to Overthink
• Not All Pixels Are Equal: Difficulty-aware Semantic Segmentation via Deep Layer Cascade
• Learning Cross-Modal Deep Representations for Robust Pedestrian Detection
• All Weather Perception: Joint Data Association, Tracking, and Classification for Autonomous Ground Vehicles
• Joint Detection and Tracking in Videos with Identification Features
• Learning Anytime Predictions in Neural Networks via Adaptive Loss Balancing
• Fused DNN: A deep neural network fusion approach to fast and robust pedestrian detection
• Adaptive Feeding: Achieving Fast and Accurate Detections by Adaptively Combining Object Detectors
• FrugalML: How to Use ML Prediction APIs More Accurately and Cheaply
• Unsupervised Hard Example Mining from Videos for Improved Object Detection
• Visual Compiler: Synthesizing a Scene-Specific Pedestrian Detector and Pose Estimator
• Exploring Multi-Branch and High-Level Semantic Networks for Improving Pedestrian Detection
• Deep Multi-camera People Detection
• A Unified Multi-scale Deep Convolutional Neural Network for Fast Object Detection
• Pushing the Limits of Deep CNNs for Pedestrian Detection
• To Boost or Not to Boost? On the Limits of Boosted Trees for Object Detection
• Which to Match? Selecting Consistent GT-Proposal Assignment for Pedestrian Detection