Human Motion Prediction via Learning Local Structure Representations and Temporal Dependencies

Authors: Xiao Guo, Jongmoo Choi2580-2587

AAAI 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Our methods achieve promising results on the Human3.6M dataset and the CMU motion capture dataset, and the code is publicly available 1.
Researcher Affiliation Academia Xiao Guo, Jongmoo Choi University of Southern California, Department of Computer Science xiaoguo@usc.edu, jongmooc@usc.edu
Pseudocode No The paper describes the architectures of Skel Net and Skel-TNet and their learning processes but does not include any formal pseudocode or algorithm blocks.
Open Source Code Yes Our methods achieve promising results on the Human3.6M dataset and the CMU motion capture dataset, and the code is publicly available 1. 1https://github.com/CHELSEA234/Skel Net motion prediction
Open Datasets Yes Human3.6M (Ionescu et al. 2014) : It has 15 different activity categories performed by professional actors (subjects) from ordinary life: walking , eating , smoking , etc. ... CMU motion capture (CMU mocap) dataset (Lab ): it contains 2235 recordings belonging to five major activity categories, being performed by 144 different subjects. ... the code is publicly available 3 .
Dataset Splits Yes Following the previous method(Martinez, Black, and Romero 2017), we use the sequences given by the subject five for testing, while rest sequences for training. ... As with (Li et al. 2018), eight actions are selected for our experiments, who are already divided into as two parts for training and testing, the code is publicly available 3 .
Hardware Specification No The paper states 'We implement our methods using the Tensorflow as the backend' but does not specify any particular CPU, GPU, or other hardware used for running the experiments.
Software Dependencies No The paper mentions 'We implement our methods using the Tensorflow as the backend' but does not provide specific version numbers for TensorFlow or any other software libraries or dependencies.
Experiment Setup Yes Regarding Skel Net, each branch consists of three linear layers, dimensions are 64, 128, 64. The last layer is of the same dimension as input, namely dimension of each time-step s mocap frame (54 and 70 dimensions for data in two datasets, respectively), denoted as input dimension. We adopt the Gradient Descent optimizer with learning rate 0.01 4. In Skel-TNet, C-RNN has a GRU with 1024 units, followed by two fully-connected layers (512, input dimension), this network is trained by Lconv, where weights ahead of Lpos, Lneg are set as 1 and 0.1. The learning rate is set as 5e-5 for Skel-TNet, with the gradient descent optimizer. Merging Network is as depicted in Figure 3, dimensions are 1024, 512, 512, input dimension, and two input weights are initialized as 0.5. We use the adam optimizer with learning rate 0.01 to train Merging Network. Throughout the entire network, LRe LUs are with the negative slope as 0.2 and dropouts are with rate 0.2 to dropout.