Let's Dance: Learning From Online Dance Videos

Abstract

In recent years, deep neural network approaches have naturally extended to the video domain, in their simplest case by aggregating per-frame classifications as a baseline for action recognition. A majority of the work in this area extends from the imaging domain, leading to visual-feature heavy approaches on temporal data. To address this issue we introduce "Let's Dance", a 1000 video dataset (and growing) comprised of 10 visually overlapping dance categories that require motion for their classification. We stress the important of human motion as a key distinguisher in our work given that, as we show in this work, visual information is not sufficient to classify motion-heavy categories. We compare our datasets' performance using imaging techniques with UCF-101 and demonstrate this inherent difficulty. We present a comparison of numerous state-of-the-art techniques on our dataset using three different representations (video, optical flow and multi-person pose data) in order to analyze these approaches. We discuss the motion parameterization of each of them and their value in learning to categorize online dance videos. Lastly, we release this dataset (and its three representations) for the research community to use.

arXiv commentfirst submitted November 2016

References in corpus (12)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• UCF101: A Dataset of 101 Human Actions Classes From Videos in The Wild
• Caffe: Convolutional Architecture for Fast Feature Embedding
• DeepPose: Human Pose Estimation via Deep Neural Networks
• The Kinetics Human Action Video Dataset
• Action Recognition with Trajectory-Pooled Deep-Convolutional Descriptors
• Towards Good Practices for Very Deep Two-Stream ConvNets
• Convolutional Two-Stream Network Fusion for Video Action Recognition
• Convolutional Pose Machines
• Sequence to Sequence -- Video to Text
• Learning Spatiotemporal Features with 3D Convolutional Networks
• Finding Action Tubes

Cited by in corpus (7)

• Fine-Grained Sports, Yoga, and Dance Postures Recognition: A Benchmark Analysis
• Music-oriented Dance Video Synthesis with Pose Perceptual Loss
• Learning Music-Dance Representations through Explicit-Implicit Rhythm Synchronization
• MOSA: Music Motion with Semantic Annotation Dataset for Cross-Modal Music Processing
• Learning Embodied Semantics via Music and Dance Semiotic Correlations
• Unsupervised 3D Pose Estimation for Hierarchical Dance Video Recognition
• Layered Neural Rendering for Retiming People in Video