Convolutional Channel Features

Abstract

Deep learning methods are powerful tools but often suffer from expensive computation and limited flexibility. An alternative is to combine light-weight models with deep representations. As successful cases exist in several visual problems, a unified framework is absent. In this paper, we revisit two widely used approaches in computer vision, namely filtered channel features and Convolutional Neural Networks (CNN), and absorb merits from both by proposing an integrated method called Convolutional Channel Features (CCF). CCF transfers low-level features from pretrained CNN models to feed the boosting forest model. With the combination of CNN features and boosting forest, CCF benefits from the richer capacity in feature representation compared with channel features, as well as lower cost in computation and storage compared with end-to-end CNN methods. We show that CCF serves as a good way of tailoring pre-trained CNN models to diverse tasks without finetuning the whole network to each task by achieving stateof-the-art performances in pedestrian detection, face detection, edge detection and object proposal generation.

Joint Face Detection and Alignment Using Multitask Cascaded Convolutional Networks

“…To evaluate the performance of our face detection method, we compare our method against the state-of-the-art methods [1,5,6,11,18,19,26,27,28,29] in FDDB, and the state-of-the-art methods [1,24,11] in WIDER FACE. Fig.…”

Pedestrian Detection Inspired by Appearance Constancy and Shape Symmetry

“…Recently, the methods based on CNN have achieved very good performance [6,18,25,26,29]. For example, Tian et al [25] proposed DeepParts to improve the detection performance by handling occlusion with an extensive part pool.…”

Survey on Artificial Intelligence for Vehicles

“…VGG16 [14] was applied to extract features, and a cascade AdaBoost classifier was trained based on these features [15,16]. Their good performance testified that CNNs have a strong power of extracting general and representative features without the need of human interference.…”