Update model card (task / checkpoint table)
Browse files
README.md
CHANGED
|
@@ -57,7 +57,7 @@ scheduler / RNG / DeepSpeed resume state is not. `code/` holds the adapter code
|
|
| 57 |
|
| 58 |
Model inputs are identical across all benchmark models: G1 state 44-D (29 body joint angles + 6 + 6 measured hand values
|
| 59 |
+ 3 projected gravity), H2 state 78-D (75 joint angles + 3 projected gravity), no joint velocities. Cameras: G1 ego view +
|
| 60 |
-
chest; H2 head + chest (
|
| 61 |
only and is never a model input. Each camera is a separate image input, except for Delta Flow and DiT4DiT, which
|
| 62 |
place the views side by side in one frame.
|
| 63 |
Training uses the Humanoid Benchmark online data augmentation spec v2 (15 photometric / geometric / sensor / lighting
|
|
|
|
| 57 |
|
| 58 |
Model inputs are identical across all benchmark models: G1 state 44-D (29 body joint angles + 6 + 6 measured hand values
|
| 59 |
+ 3 projected gravity), H2 state 78-D (75 joint angles + 3 projected gravity), no joint velocities. Cameras: G1 ego view +
|
| 60 |
+
chest; H2 head + chest (the 3-camera re-trains of 2026-11-12/13 also used the right-wrist camera; the `protocol.json` of each folder lists its cameras). The third-person camera recorded in the datasets is for data review
|
| 61 |
only and is never a model input. Each camera is a separate image input, except for Delta Flow and DiT4DiT, which
|
| 62 |
place the views side by side in one frame.
|
| 63 |
Training uses the Humanoid Benchmark online data augmentation spec v2 (15 photometric / geometric / sensor / lighting
|