Data underlying the MSc Thesis: Toward Occlusion Capable Human Trajectory Prediction
doi: 10.4121/3dc88884-d8f4-42db-b643-e799fe7fb432
Here are the dataset and model files related to the MSc thesis: Toward Occlusion Capable Human Trajectory Prediction.
This thesis focuses on handling occlusions and partially missing positional information of agents when predicting their trajectories.
Dataset files comprise three distinct versions of the trajectory dataset used throughout the thesis project:
- train, val and test splits for the trajectory dataset under fully observed conditions
- train, val and test splits for the trajectory dataset under occluded conditions
- test split for the trajectory dataset under occluded conditions, with imputation of missing positions by means of interpolation and constant velocity extrapolation.
Occluded trajectories were generated by applying a simulator of occlusion events onto a publicly available trajectory dataset: the Stanford Drone Dataset. Our dataset Files are therefore derived from the Stanford Drone Dataset.
Model files contain checkpoints with weights that can be used to initialize prediction models from our implementation. These checkpoints are accompanied by some metadata, with information about the evolution of train and validation losses throughout the training process of individual model instances. Models are trained in two separate phases (I, and II): each model file contains all relevant model data for both phases of one model instance.
The source code related to these files is hosted on this GitHub page. The repository's README contains a comprehensive set of instructions on how to use the files in order to reproduce the results we obtained in our research (please refer to the section titled "Downloading Models and Legacy Datasets").
- 2025-01-24 first online, published, posted
DATA
- 21,045 bytesMD5:
969ab2eb3f7a1f98600e3b480a1e5f94
README.md - 78,662,265 bytesMD5:
9a35bfefe45e6a5b258271e0fa21ff44
agentformer_100.tar.gz - 78,309,483 bytesMD5:
114e8daa52178ef80a2b6f53bc57a949
agentformer_101.tar.gz - 77,952,509 bytesMD5:
5df3f20b3c40c6216f1ef52109a611da
agentformer_102.tar.gz - 78,380,864 bytesMD5:
03cbceaf5982dbc62d2ea63c74d728cc
agentformer_103.tar.gz - 78,555,217 bytesMD5:
8a04ca820dbf2866f669209c6c5e08f9
agentformer_104.tar.gz - 188,036,682 bytesMD5:
9b69fa6c02c55210b94422cc12394b79
fully_observed.tar.gz - 329,145,349 bytesMD5:
3f96f1c1282e013ae8c123c96dfbb647
occlusion_simulation.tar.gz - 38,525,982 bytesMD5:
363d96c3b15380d7fd9a61c6784ee572
occlusion_simulation_imputed.tar.gz - 77,933,679 bytesMD5:
7f5610053b30f809cff89d4a6451b48e
occlusionformer_DS.tar.gz - 103,608,495 bytesMD5:
19067e14c35ca015481da18d13804eb7
occlusionformer_DS_mapA.tar.gz - 103,287,342 bytesMD5:
0b4b4e296f99f8f501805cd20ee5a566
occlusionformer_DS_mapB.tar.gz - 78,129,858 bytesMD5:
26f9b7936a65c9f6fe68545803ae96de
occlusionformer_FO_1.tar.gz - 78,049,542 bytesMD5:
11d93607a0dabb2f43edab178318b024
occlusionformer_FO_2.tar.gz - 77,993,903 bytesMD5:
8b4b1e4499c0d728db70a54c396c76bb
occlusionformer_FO_3.tar.gz - 77,866,733 bytesMD5:
aaa2b9c46148c479f12fb544cecd7b30
occlusionformer_FO_4.tar.gz - 77,632,121 bytesMD5:
1b4d6eb65ac19b2271d8d7bd624d6a13
occlusionformer_FO_5.tar.gz - 77,334,853 bytesMD5:
f5f2804a9a097320d074dea421dc4b15
occlusionformer_OS.tar.gz -
download all files (zip)
1,699,425,922 bytes unzipped