What Can Help Pedestrian Detection?

By Jiayuan Mao, Tete Xiao, Yuning Jiang, Zhimin Cao. 2017

293 citations recorded by the source corpus.

Aggregating extra features has been considered as an effective approach to boost traditional pedestrian detection methods. However, there is still a lack of studies on whether and how CNN-based pedestrian detectors can benefit from these extra features. The first contribution of this paper is exploring this issue by aggregating extra features into CNN-based pedestrian detection framework. Through extensive experiments, we evaluate the effects of different kinds of extra features quantitatively. Moreover, we propose a novel network architecture, namely HyperLearner, to jointly learn pedestrian detection as well as the given extra feature. By multi-task training, HyperLearner is able to utilize the information of given features and improve detection performance without extra inputs in inference. The experimental results on multiple pedestrian benchmarks validate the effectiveness of the proposed HyperLearner.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Jiashi Feng
2. Dacheng Tao
3. Songtao Liu
4. Gang Yu
5. Zhiyuan Liu
6. Jian‐Yuan Sun
7. Maosong Sun
8. Steven C. H. Hoi

In Computer Vision and Pattern Recognition

1. Dacheng Tao
2. Luc Van Gool
3. Xiaogang Wang
4. Ming-Hsuan Yang
5. Ling Shao
6. Jiashi Feng
7. Yi Yang
8. Trevor Darrell

Citations in the corpus

• MegDet: A Large Mini-Batch Object Detector (2017, methodology): “Based on our MegDet, we integrate the techniques including OHEM [35], atrous convolution [40, 2], stronger base models [38, 18], large kernel [28], segmentation supervision [27, 34], diverse network structure [12, 32, 36], contextual modules [22, 9], ROIAlign [14] and multi-scale training and testin”
• Pedestrian Detection: The Elephant In The Room (2020, background): “After the initial works, Faster RCNN [30] became most popular framework with wide range of literature deploying it for pedestrian detection [44,42,6,5,25].”
• Object Detection Algorithm Based on Improved YOLOv3 (2020, methodology): “It has been widely used in autonomous driving [2], pedestrian detection [3], medical imaging [4], industrial detection [5], robot vision [6], intelligent video surveillance [7], remote sensing images [8], etc.”
• Retina U-Net: Embarrassingly Simple Exploitation of Segmentation Supervision for Medical Object Detection (2018, methodology): “While previous work in the non-medical domain has pursued this concept [23, 19, 20, 3], the extra supervision signal is either used sub-optimally or combined with approaches of arguably significant model complexity.”
• Recent Advances in Deep Learning for Object Detection (2019, background): “[215] claimed that properly aggregating extra features into pedestrian detector can boost detection accuracy.”
• EuroCity Persons: A Novel Benchmark for Person Detection in Traffic Scenes (2018, methodology): “For Faster R-CNN, R-FCN and SSD we use VGG-16 [45] as base network as it is very common in pedestrian detection [4], [60], [62], [63], [71].”
• Triply Supervised Decoder Networks for Joint Detection and Segmentation (2018, methodology): “ segmentation are in parallel attached to the last layers of the encoder [2]. In Fig. 1(b), the branch for object detection is further refined by the features from the branch for semantic segmentation [31, 47]. Recently, the encoder-decoder network is further used for joint object de1 arXiv:1809.0929”
• Adaptive NMS: Refining Pedestrian Detection in a Crowd (2019, background): “[26] presents a jointly learning framework with extra features to further improve performance.”
• Spatial Attentive Single-Image Deraining With a High Quality Real Rain Dataset (2019, background): “, pedestrian detection [30], visual tracking [37], or road sign recognition [48].”
• Temporal-Context Enhanced Detection of Heavily Occluded Pedestrians (2020, background): “With the renaissance of convolutional neural networks, many deep learning based methods on pedestrian detection [27, 28, 10, 26, 24, 23, 19, 17, 25, 18, 48, 36] significantly outperform the hand-crafted feature based methods [55, 61, 21, 14].”
• Occlusion-aware R-CNN: Detecting Pedestrians in a Crowd (2018, methodology): “Similar to [55], we evaluate the OR-CNN method on the Reasonable subset of the Caltech-USA dataset, and compare it to other state-of-the-art methods ( e.g. , [55,66,50,5,6,63,30,25,49,9,34,15]) in Figure 4.”
• Coupled End-to-End Transfer Learning with Generalized Fisher Information (2018, background): “It is decoded in the reconstruction T1 ◦ S(x), which is a resemblance to the channel features in [27].”
• Legal Judgment Prediction via Topological Learning (2018, background): “It can transfer useful information among various tasks and has been applied to a wide range of areas, including NLP (Collobert and Weston, 2008), speech recognition (Deng et al., 2013), and computer vision (Girshick, 2015; Mao et al., 2017).”
• MetaAnchor: Learning to Detect Objects with Customized Anchors (2018, background): “First, anchor boxes need careful design, e.g. via clustering [30], which is especially critical on specific detection tasks such as anchor-based face [39, 44, 27, 35, 42] and pedestrian [40, 3, 43, 25] detections.”
• Object Detection in 20 Years: A Survey (2019, background): “pedestrians due to the low resolution of their convolutional features [165]. Some recent solutions to this problem include feature fusion [165], introducing extra high-resolution handcrafted features [313, 314], and ensembling detection results on multiple resolutions [315]. To improve hard negative”
• CrowdHuman: A Benchmark for Detecting Human in a Crowd (2018, background): “[18] propose a multi-task network to further improve detection performance.”
• Fusion of Multispectral Data Through Illumination-aware Deep Neural Networks for Pedestrian Detection (2018, background): “[26] proposed a novel network architecture to jointly learn pedestrian detection as well as the given extra feature.”
• Bi-box Regression for Pedestrian Detection and Occlusion Estimation (2018, methodology): “In [5, 37, 20, 41, 34, 4], Fast R-CNN [16] or Faster R-CNN [27] is adapted for pedestrian detection.”
• High-Level Semantic Feature Detection: A New Perspective for Pedestrian Detection (2019, methodology): “Following [51, 32, 46, 28, 52], we use the training data augmented by 10 folds (42782 frames) and test on the 4024 frames in the standard test set, all experiments are conducted on the new annotations provided by [49].”