Add files using upload-large-folder tool

README.md

@@ -0,0 +1,48 @@
+---
+license: cc-by-nc-sa-4.0
+language:
+- en
+tags:
+- human-trajectory-prediction
+- jrdb
+- jta
+- physics-simulation
+- paa2026
+pipeline_tag: other
+library_name: pytorch
+---
+
+# Hexa-P: PAA 2026 release assets
+
+Checkpoints and preprocessed data for
+**Hexa-P: Physical Plausibility-aware Pre-training for Pedestrian Path Prediction** (Pattern Analysis and Applications, 2026).
+
+🐙 [code](https://github.com/ImIntheMiddle/Hexa-P) · 🌐 [project page](https://iminthemiddle.github.io/Hexa-P-Page/)
+
+## Contents
+
+- `weights/stage1/hexap_pretrain.ckpt`: Hexa-P pre-trained in the physics simulation (5 hypotheses), the starting point of fine-tuning
+- `weights/stage1/hexap_pretrain_deterministic.ckpt`: the same with one hypothesis
+- `weights/stage1/controller.ckpt`: humanoid controller that walks the pre-training paths
+- `weights/stage2/{jrdb, jta}/hexap.pt`: Hexa-P fine-tuned on JRDB and JTA
+- `data/{jta, jrdb, vru}/{train, val, test}/`: preprocessed data (`vru` is the Urban dataset)
+- `data/jta_replay/test/`: JTA test scenes for the robot navigation experiment
+
+## Usage
+
+See the [GitHub README](https://github.com/ImIntheMiddle/Hexa-P#weights-and-data) for download + setup instructions.
+
+## License
+
+Research, non-commercial use only. `data/<dataset>/LICENSE` gives the licence of each folder: JTA CC BY-NC 4.0, JRDB CC BY-NC-SA 4.0, Urban CC BY 4.0. The weights were trained with SMPL and AMASS, which restrict commercial use. The JTA authors require the statement: "I declare that I will use the JTA Dataset for research and educational purposes only, since I am aware that commercial use is prohibited. I also undertake to purchase a copy of Grand Theft Auto V."
+
+## Citation
+
+```bibtex
+@article{taketsugu2026hexap,
+  title   = {Hexa-P: Physical Plausibility-aware Pre-training for Pedestrian Path Prediction},
+  author  = {Taketsugu, Hiromu and Nobuhara, Shohei and Ukita, Norimichi},
+  journal = {Pattern Analysis and Applications},
+  year    = {2026}
+}
+```

data/jrdb/LICENSE

@@ -0,0 +1,11 @@
+Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0)
+https://creativecommons.org/licenses/by-nc-sa/4.0/legalcode
+
+Derived from JRDB (CC BY-NC-SA 3.0): R. Martin-Martin et al., "JRDB: A Dataset and Benchmark of Egocentric
+Robot Visual Perception of Humans in Built Environments", TPAMI 2021, and S. Ehsanpour et al., "JRDB-Act",
+CVPR 2022, via the preprocessed data of Social-Transmotion (S. Saadatnejad et al., ICLR 2024).
+Changes: 3D body poses regressed from the JRDB images with BEV and added to the trajectories, joints
+remapped to the body order of the SMPL humanoid, windows packed into .pt files. meta.pt lists the scene,
+frame and person ids of each window.
+
+Provided as is, without warranty, and not endorsed by the authors of JRDB.

data/jta/LICENSE

@@ -0,0 +1,12 @@
+Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0)
+https://creativecommons.org/licenses/by-nc/4.0/legalcode
+
+Derived from JTA: M. Fabbri et al., "Learning to Detect and Track Visible and Occluded Body Joints in a
+Virtual World", ECCV 2018, via the preprocessed data of Social-Transmotion (S. Saadatnejad et al., ICLR 2024).
+Changes: joints remapped to the body order of the SMPL humanoid; windows packed into .pt files.
+
+The JTA authors require the statement: "I declare that I will use the JTA Dataset for research and
+educational purposes only, since I am aware that commercial use is prohibited. I also undertake to
+purchase a copy of Grand Theft Auto V."
+
+Provided as is, without warranty, and not endorsed by the authors of JTA.

data/jta/train/part_15.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8ad4229ce2cf92b18099db2e93acbe5c5f1dc9e027b92f76f12a49231a09d4e2
+size 353241250

data/jta_replay/LICENSE

@@ -0,0 +1,12 @@
+Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0)
+https://creativecommons.org/licenses/by-nc/4.0/legalcode
+
+Derived from JTA: M. Fabbri et al., "Learning to Detect and Track Visible and Occluded Body Joints in a
+Virtual World", ECCV 2018, via the preprocessed data of Social-Transmotion (S. Saadatnejad et al., ICLR 2024).
+Changes: joints remapped to the body order of the SMPL humanoid; whole test sequences packed into replay scenes.
+
+The JTA authors require the statement: "I declare that I will use the JTA Dataset for research and
+educational purposes only, since I am aware that commercial use is prohibited. I also undertake to
+purchase a copy of Grand Theft Auto V."
+
+Provided as is, without warranty, and not endorsed by the authors of JTA.

data/vru/LICENSE

@@ -0,0 +1,9 @@
+Creative Commons Attribution 4.0 International (CC BY 4.0)
+https://creativecommons.org/licenses/by/4.0/legalcode
+
+Derived from "Pedestrians and Cyclists in Road Traffic: Trajectories, 3D Poses and Semantic Maps" by
+V. Kress et al. (https://doi.org/10.5281/zenodo.4898838), licensed under CC BY 4.0.
+Changes: 4 observed and 12 future frames at 5 fps sampled from the pedestrian recordings, joints remapped
+to the body order of the SMPL humanoid, windows packed into .pt files.
+
+Provided as is, without warranty, and not endorsed by the authors of the dataset.

data/vru/train/part_26.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d88f7fbe137c9534ef84ba29d2161959bceb329cfc015aef2d26abc16771cfcd
+size 45247930

data/vru/train/part_31.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a518dcde316ee172a2b407f3781986d32cf65053d39958c82ad81ae3d52c9cd8
+size 45247930

data/vru/train/part_32.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:942afe96b565bfb74c2f4f4573ef3fd5175ae03d044057be4581ff187b8810f0
+size 45247930

data/vru/train/part_33.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b06ffd1c85548b0057f8a9bd7d74e45fd705668d7b643c8bcf75a48c609efd8d
+size 45247930

data/vru/train/part_35.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f60df068b44ff75b1f388b07101116fb2dd10fd9a4ca4ef9ea86983083ae663c
+size 35089338

data/vru/train/part_5.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:81dd3bf0fd77a7d3ae6b93c08f4781344a462978e63a5779addf5579bd0d5a44
+size 45227670

data/vru/train/part_6.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:33f89d7eea65f1ceb137cca3b4610f12bfa90ca0110a377085b9da2cecdcea75
+size 45227670

data/vru/train/part_9.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:44cd198aa0e0fd740527206789b84d915705a9b9da31d48f9635125bee684ec7
+size 45227670

data/vru/val/part_10.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2f92b53f4e4c34a04ec24cb9cdb69e429a9c86aa2996871d6bbab603a140a5e2
+size 8695620