2019 · 19 citations · 16 references
ComputingEducational TechnologyComputer VisionMovement AnalysisSensor FusionAction RecognitionGesture AnalysisGesture RecognitionMultimodal InteractionPose EstimationClassroom Behavior AnalysisGesture PosesGesture SensingHand Gesture AnalysisHand GesturesHand Keypoint DetectionInformation Fusion NetworksPose Estimation For Gestures
With the development of visual technology, automatic classroom behavior analysis has attracted more and more attention, hand-raising gesture recognition is the most essential one of them. There exist a lot of challenging cases in the uncontrolled environment, such as unknown number of students and their position, occluded body parts and low-resolution hands area. In order to solve the above challenges, we propose a two-stage method for identifying hand-raising gesture in real classroom scenarios. The first stage performs a multi-stage pose estimation to get the candidate hand-based area of each student, only assemble keypoints on the raising arm and use multi-scale information fusion to get finer results. The second stage identify the gesture style through binary classi-fication network. The experiment has proved the effectiveness of our proposed method and achieved an accuracy of 94.76% with our self-built hand-raising dataset.
16
Deep Residual Learning for Image Recognition
Kaiming He, Xiangyu Zhang, Shaoqing Ren et al. · 2016 · 222.1K citations · Full text
Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
Shaoqing Ren, Kaiming He, Ross Girshick et al. · arXiv (Cornell University) · 2015 · 18.2K citations · Full text
Realtime Multi-person 2D Pose Estimation Using Part Affinity Fields
Zhe Cao, Tomas Simon, Shih-En Wei et al. · 2017 · 7.3K citations
R-FCN: Object Detection via Region-based Fully Convolutional Networks
Jifeng Dai, Yi Li, Kaiming He et al. · arXiv (Cornell University) · 2016 · 3.4K citations · Full text
2D Human Pose Estimation: New Benchmark and State of the Art Analysis
Mykhaylo Andriluka, Leonid Pishchulin, Peter Gehler et al. · 2014 · 2.8K citations