Bi-box Regression for Pedestrian Detection and Occlusion Estimation

Zhou, Chunluan; Yuan, Junsong

doi:10.1007/978-3-030-01246-5_9

Cited by 201 publications

(181 citation statements)

References 36 publications

Supporting

Mentioning

178

Contrasting

Unclassified

Order By: Relevance

“…For training with R+ set, we compare with Bi-Box [26] and MGAN [33]. Similar to HBAN, Bi-Box also adopts a parallel branches detection framework.…”

Section: Methodsmentioning

confidence: 99%

“…However, as we focus on detecting pedestrians with various occlusion levels, different subsets are used in the experiments to train HBAN in order to provide a thorough evaluation. Following [26,33], we also use another training set which contains pedestrians with occlusion ratio less than 70%, denoted as R+. Training Details.…”

Section: Implementation Settingsmentioning

confidence: 99%

“…We train for 20k iterations first with the base learning rate of 0.001 and another 10k iterations with learning rate decreased by 10. When training on R+ set, we also adopt visible ratio to sample positive RoIs, as introduced in Bi-Box [26]. The SGD solver and a mini-batch of 2 images is adopted to optimize the parameters on a single Titan X GPU.…”

Section: Implementation Settingsmentioning

confidence: 99%

“…But as we focus on pedestrians with various occlusion levels, training on R is insufficient to thoroughly evaluate our model. Therefore, following Bi-Box [26], we use another training set, which contains pedestrians with height ≥ 50 and occlusion ≤ 0.7, denoted as R+. The results are demonstrated in Table 2 and Table 3, respectively.…”

Section: Ablation Studiesmentioning

confidence: 99%

See 3 more Smart Citations

Semantic head enhanced pedestrian detection in a crowd

Wang

2020

Neurocomputing

View full text Add to dashboard Cite

Pedestrian detection in the crowd is a challenging task because of intra-class occlusion. More prior information is needed for the detector to be robust against it. Human head area is naturally a strong cue because of its stable appearance, visibility and relative location to body. Inspired by it, we adopt an extra branch to conduct semantic head detection in parallel with traditional body branch. Instead of manually labeling the head regions, we use weak annotations inferred directly from body boxes, which is named as 'semantic head'. In this way, the head detection is formulated into using a special part of labeled box to detect the corresponding part of human body, which surprisingly improves the performance and robustness to occlusion. Moreover, the head-body alignment structure is explicitly explored by introducing Alignment Loss, which functions in a self-supervised manner. Based on these, we propose the head-body alignment net (HBAN) in this work, which aims to enhance pedestrian detection by fully utilizing the human head prior. Comprehensive evaluations are conducted to demonstrate the effectiveness of HBAN on CityPersons dataset.

show abstract

“…For training with R+ set, we compare with Bi-Box [26] and MGAN [33]. Similar to HBAN, Bi-Box also adopts a parallel branches detection framework.…”

Section: Methodsmentioning

confidence: 99%

Section: Implementation Settingsmentioning

confidence: 99%

Section: Implementation Settingsmentioning

confidence: 99%

Section: Ablation Studiesmentioning

confidence: 99%

See 2 more Smart Citations

Semantic head enhanced pedestrian detection in a crowd

Wang

2020

Neurocomputing

View full text Add to dashboard Cite

show abstract

“…Previous work about visionbased pedestrian protection systems [1] provides a thorough investigation of such methods based on shallow learning. Recently, various deep learning methods are proposed for singlestage detection [5], [6], detection in a crowd [7], [8], and detection at the presence of occlusion [5], [25], [26]; all these methods obtain prominent accuracies for pedestrian detection. For pedestrian tracking, multi-person tracking methods [27], [28] are proposed to track every person in a crowded scene.…”

Section: Related Workmentioning

confidence: 99%

Spatiotemporal Relationship Reasoning for Pedestrian Intent Prediction

Liu

Adeli

Cao

et al. 2020

IEEE Robot. Autom. Lett.

141

View full text Add to dashboard Cite

Reasoning over visual data is a desirable capability for robotics and vision-based applications. Such reasoning enables forecasting the next events or actions in videos. In recent years, various models have been developed based on convolution operations for prediction or forecasting, but they lack the ability to reason over spatiotemporal data and infer the relationships of different objects in the scene. In this paper, we present a framework based on graph convolution to uncover the spatiotemporal relationships in the scene for reasoning about pedestrian intent. A scene graph is built on top of segmented object instances within and across video frames. Pedestrian intent, defined as the future action of crossing or not-crossing the street, is very crucial piece of information for autonomous vehicles to navigate safely and more smoothly. We approach the problem of intent prediction from two different perspectives and anticipate the intention-to-cross within both pedestrian-centric and location-centric scenarios. In addition, we introduce a new dataset designed specifically for autonomousdriving scenarios in areas with dense pedestrian populations: the Stanford-TRI Intent Prediction (STIP) dataset. Our experiments on STIP and another benchmark dataset show that our graph modeling framework is able to predict the intention-to-cross of the pedestrians with an accuracy of 79.10% on STIP and 79.28% on Joint Attention for Autonomous Driving (JAAD) dataset up to one second earlier than when the actual crossing happens. These results outperform baseline and previous work. Please refer to http://stip.stanford.edu/ for the dataset and code.Index Terms-spatiotemporal graphs, forecasting, graph neural networks, autonomous-driving. Recent work [19]-[23] introduced pedestrian intent prediction and have typically tackled the problem by observing pedestrian-specific features such as location, velocity, and

show abstract

Feature Enhancement for Joint Human and Head Detection

Zhang

Zhuang

et al. 2019

Biometric Recognition

View full text Add to dashboard Cite

Human and head detection have been rapidly improved with the development of deep convolutional neural networks. However, these two detection tasks are often studied separately, without taking advantage of the relationship between human and head. In this paper, we present a new two-stage detection framework, namely Joint Enhancement Detection (JED), to simultaneously detect human and head based on enhanced features. Specifically, the proposed JED contains two newly added modules, i.e., the Body Enhancement Module (BEM) and the Head Enhancement Module (HEM). The former is designed to enhance the features used for human detection, while the latter aims to enhance the features used for head detection. With these enhanced features in a joint framework, the proposed method is able to detect human and head simultaneously and efficiently. We verify the effectiveness of the proposed method on the CrowdHuman dataset and achieve better performance than baseline method for both human and head detection.

show abstract

Bi-box Regression for Pedestrian Detection and Occlusion Estimation

Cited by 201 publications

References 36 publications

Semantic head enhanced pedestrian detection in a crowd

Semantic head enhanced pedestrian detection in a crowd

Spatiotemporal Relationship Reasoning for Pedestrian Intent Prediction

Feature Enhancement for Joint Human and Head Detection

Contact Info

Product

Resources

About