Video Prediction via Example Guidance

Authors: Jingwei Xu, Huazhe Xu, Bingbing Ni, Xiaokang Yang, Trevor Darrell

ICML 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We conduct extensive experiments on several widely used datasets, including moving digit (Srivastava et al., 2015), robot arm motion (Finn et al., 2016), and human activity (Zhang et al., 2013). Considerable enhancement is observed both in quantitative and qualitative aspects.
Researcher Affiliation Academia 1Shanghai Jiao Tong University 2University of California, Berkeley. Correspondence to: Bingbing Ni <nibingbing@sjtu.edu.cn>.
Pseudocode Yes Algorithm 1 Example Guided Video Prediction
Open Source Code Yes Project Page: https://sites.google.com/view/vpeg-supp/home.
Open Datasets Yes We evaluate our model with three widely used video prediction datasets: (1) Moving Mnist (Srivastava et al., 2015), (2) Bair Robot Push (Ebert et al., 2017) and (3) Penn Action (Zhang et al., 2013).
Dataset Splits No The paper references datasets and mentions 'training' and 'testing' phases but does not explicitly provide specific train/validation/test dataset splits (e.g., percentages, sample counts, or explicit standard split citations).
Hardware Specification No The paper does not provide specific hardware details (e.g., GPU/CPU models, memory specifications) used for running the experiments.
Software Dependencies No The paper discusses various models and architectures (e.g., Conv-LSTM, GAN, VAE, LSTM) but does not provide specific version numbers for any software dependencies or libraries used for implementation.
Experiment Setup Yes In all experiments we empirically set N = K = 5.