Aggregated Multi-GANs for Controlled 3D Human Motion Prediction

Authors: Zhenguang Liu, Kedi Lyu, Shuang Wu, Haipeng Chen, Yanbin Hao, Shouling Ji2225-2232

AAAI 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experiments show that our method outperforms state-of-the-art.
Researcher Affiliation Academia Zhenguang Liu1, Kedi Lyu2 , Shuang Wu3*, Haipeng Chen2*, Yanbin Hao4, Shouling Ji5 1Zhejiang Gongshang University 2Jilin University 3Nanyang Technological University 4University of Science and Technology of China 5Zhejiang University
Pseudocode No The paper describes the architecture and loss functions in detail but does not include any pseudocode or algorithm blocks.
Open Source Code Yes The codes are available at https://github.com/herolvkd/AM-GAN.
Open Datasets Yes In order to verify the ability of our model, we selected the largest and widely used data set: H3.6m (Ionescu et al. 2014) that contains 3.6 million human images with 3D poses (comprising 25 distinct skeletal joints) obtained via the Vicon mocap system. ... subject 5 is used for testing while the data for the other 6 subjects are used for training.
Dataset Splits No The paper specifies a training and testing split for the H3.6m dataset (subject 5 for testing, others for training) but does not mention a separate validation split or how validation was performed.
Hardware Specification No The paper does not specify the hardware used for experiments (e.g., CPU, GPU models, memory, or specific computing cluster details).
Software Dependencies No The paper states: "We use Tensorflow to implement our method." However, it does not provide a version number for TensorFlow or any other software dependencies.
Experiment Setup Yes The encoder and decoder of VAE consist of a GRU with 1,024 hidden units. ... The aggregation module is a full connected layer with 1,024 units. The local critic is a three-layer fully connected feedforward network with 512 units, while the global critic possesses 1,024 units. The Adam Optimizer (Kingma and Ba 2014) is employed with initial learning 5e-5 and the batch size is set to 16.