Conference Paper

Detecting Pedestrians by Learning Shapelet Features.

DOI: 10.1109/CVPR.2007.383134 Conference: 2007 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR 2007), 18-23 June 2007, Minneapolis, Minnesota, USA
Source: DBLP

ABSTRACT In this paper, we address the problem of detecting pedes- trians in still images. We introduce an algorithm for learn- ing shapelet features, a set of mid-level features. These fea- tures are focused on local regions of the image and are built from low-level gradient information that discriminates be- tween pedestrian and non-pedestrian classes. Using Ad- aBoost, these shapelet features are created as a combina- tion of oriented gradient responses. To train the final classi- fier, we use AdaBoost for a second time to select a subset of our learned shapelets. By first focusing locally on smaller feature sets, our algorithm attempts to harvest more use- ful information than by examining all the low-level features together. We present quantitative results demonstrating the effectiveness of our algorithm. In particular, we obtain an error rate 14 percentage points lower (at10−6 FPPW) than the previous state of the art detector of Dalal and Triggs (1) on the INRIA dataset.

  • Source
    • "It requires target feature description from input space. Existing feature description methods are: symmetry and edge density characteristics [5], Viola's Haar-like features [6], SIFT descriptor proposed by Lowe [7], and Shapelet features by Sbazmeydani [8]. Most of all, Dalal and Triggs [9] proposed the Histogram of Gradient feature (HOG) descriptor, which proves to be effective in human detection. "
  • Source
    • "The first one is related to the features used to represent pedestrians. Haarlike features [40], shapelets [36], shape context [26] and histogram of oriented gradient (HOG) [7] features are commonly used. The last one is the most popular and almost all detectors use it in some forms. "
    [Show abstract] [Hide abstract]
    ABSTRACT: The importance of pedestrian detection in many applications has led to the development of many algorithms. In this paper, we address the problem of combining the outputs of several detectors. A pre-trained pedestrian detector is seen as a black box returning a set of bounding boxes with associated scores. A calibration step is first conducted to transform those scores into a probability measure. The bounding boxes are then grouped into clusters and their scores are combined. Different combination strategies using the theory of belief functions are proposed and compared to probabilistic ones. A combination rule based on triangular norms is used to deal with dependencies among detectors. More than 30 state-of-the-art detectors were combined and tested on the Caltech Pedestrian Detection Benchmark. The best combination strategy outperforms the currently best performing detector by 9% in terms of log-average miss rate.
    25th British Machine Vision Conference, Nottingham, UK; 09/2014
  • Source
    • "Later, [12] proposed a human classification algorithm that uses Histogram of Oriented Gradients (HOG) relying on the dense representation of histogram of gradient within a detection window. Similarly, gradient histogram based feature extraction methods can be found in [13], [14], and (EOH) [15]. HOG achieved a promising result on human detection and was popularly used in combination with texture and color features to further improve the accuracy as Local Binary Pattern (LBP) [16], and Local Ternary Pattern [17]. "
    Multimedia and Expo (ICME), 2014 IEEE International Conference on, Chen Du, China; 07/2014
Show more


Available from