Update README.md

README.md

@@ -1,3 +1,80 @@
----
-license: apache-2.0
----
+---
+license: apache-2.0
+---
+
+# ACT-Estimator
+
+This model is designed for use with [ACT-Bench](https://github.com/turingmotors/ACT-Bench) to compute evaluation metrics. It serves as a prediction model that reconstructs trajectories from driving videos generated by autonomous driving world models.
+The predicted trajectory is compared against the instruction trajectory, which serves as the reference, to calculate the evaluation metrics for ACT-Bench.
+
+
+## Model Summary
+
+- Developed by: Turing Inc.
+- Licence: Apache License 2.0
+- Model Size: 20.4M
+
+
+## Model Date
+
+`ACT-Estimator` was trained on November 2024.
+
+
+## Model I/O
+
+Input
+- `generated_videos` (shape: (batch_size, 3, 44, 224, 224))
+- `timestamps` (shape: (batch_size, 44))
+
+Output
+- `command` (shape: (batch_size, 9))
+- `waypoints` (shape: (batch_size, 44, 2))
+
+## Training Dataset
+
+- Video frames-trajectory pairs from [nuScenes](https://www.nuscenes.org/) dataset. Details are described in our [paper]().
+
+
+## Authors
+
+Here are the team members who contributed to the development of `ACT-Bench`:
+- Hidehisa Arai
+- Keishi Ishihara
+- Tsubasa Takahashi
+- Yu Yamaguchi
+
+
+## How to use
+
+```python
+import torch
+from transformers import AutoModel
+
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+
+act_estimator = AutoModel.from_pretrained("turing-motors/Act-Estimator", trust_remote_code=True)
+act_estimator.to(device)
+act_estimator.eval()
+
+# dummy inputs
+generated_videos = torch.randn(1, 3, 44, 224, 224).to(device)
+timestamps = torch.randn(1, 44).to(device)
+
+out = act_estimator(generated_videos, timestamps)
+out.keys()
+# dict_keys(['command', 'waypoints'])
+```
+
+
+## License
+
+The ACT-Estimator is licensed under the Apache License 2.0.
+
+
+## Citation
+
+If you find our work helpful, please feel free to cite us.
+
+```
+coming soon
+```