Representing Spatial Trajectories as Distributions

Authors: Didac Suris Coll-Vinent, Carl Vondrick

NeurIPS 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experiments show our method s advantage over baselines in prediction tasks. Our experiments on human movement datasets show that our method can accurately predict the past and future of a trajectory segment, as well as the interpolation between two different segments, outperforming autoregressive baselines. Additionally, it can do so for any continuous point in time.
Researcher Affiliation Academia Dídac Surís Columbia University didac.suris@columbia.edu Carl Vondrick Columbia University vondrick@cs.columbia.edu
Pseudocode No The paper describes the model architecture and training process but does not include explicit pseudocode or algorithm blocks.
Open Source Code Yes See trajectories.cs.columbia.edu for video results and code. Yes, they are in the supplementary materials.
Open Datasets Yes We extract human movement trajectories from the Fine Gym [45], Diving48 [29] and Fis V [57] datasets, which correspond to gymnastics, diving and figure skating, respectively.
Dataset Splits No The paper mentions training and testing, and refers to Appendix B and C for more details, but it does not explicitly state specific training, validation, or test dataset split percentages or sample counts in the main text.
Hardware Specification No The paper states that training details are in Appendix C, but the main text does not specify any particular hardware (e.g., GPU models, CPU types) used for the experiments.
Software Dependencies No The paper mentions using a Transformer Encoder, ResNet, OpenPose, and cites PyTorch, but it does not provide specific version numbers for any of these software components.
Experiment Setup No The paper describes the general architecture and some training concepts (e.g., triplet loss, reparameterization trick, using box embeddings), but it refers to Appendix C for more details and does not provide specific hyperparameter values like learning rates, batch sizes, or optimizer settings in the main text.