# Appendix B.3 — Visual task examples

**48 full-size case figures · 20 task families · 3 benchmark levels**

These are the complete figures prepared for Appendix B.3. Each task page groups its input-modality variants. Click a figure to open its full-size PNG, or use the original PDF link for zooming and download.

The cards explain benchmark inputs and motion assets; they are not model-generated predictions. For Level 2, read the target/reference wording printed in each figure: some tasks show an original reference motion rather than a computed modified result.

[Back to the RoboSteer dataset](https://huggingface.co/datasets/YanCORANV/RoboSteer) · [Task media examples](https://huggingface.co/datasets/YanCORANV/RoboSteer/blob/main/docs/examples.md)

## Level 1

### Full conditioning

- [Text-to-Motion Generation](https://huggingface.co/datasets/YanCORANV/RoboSteer/blob/main/docs/appendix-b3/text-to-motion-generation.md) — 1 figure
- [Video-to-Motion Imitation](https://huggingface.co/datasets/YanCORANV/RoboSteer/blob/main/docs/appendix-b3/video-to-motion-imitation.md) — 1 figure
- [Audio-to-Motion Generation](https://huggingface.co/datasets/YanCORANV/RoboSteer/blob/main/docs/appendix-b3/audio-to-motion-generation.md) — 1 figure
- [Rhythm-to-Motion Alignment](https://huggingface.co/datasets/YanCORANV/RoboSteer/blob/main/docs/appendix-b3/rhythm-to-motion-alignment.md) — 1 figure
- [Rotation-to-Pose Generation](https://huggingface.co/datasets/YanCORANV/RoboSteer/blob/main/docs/appendix-b3/rotation-to-pose-generation.md) — 1 figure
### Temporal completion

- [Motion Prediction](https://huggingface.co/datasets/YanCORANV/RoboSteer/blob/main/docs/appendix-b3/motion-prediction.md) — 5 figures
- [Motion Retrodiction](https://huggingface.co/datasets/YanCORANV/RoboSteer/blob/main/docs/appendix-b3/motion-retrodiction.md) — 4 figures
- [Motion Interpolation](https://huggingface.co/datasets/YanCORANV/RoboSteer/blob/main/docs/appendix-b3/motion-interpolation.md) — 4 figures
- [Key-frame Conditioning](https://huggingface.co/datasets/YanCORANV/RoboSteer/blob/main/docs/appendix-b3/key-frame-conditioning.md) — 4 figures
### Spatial completion

- [Upper-to-Full Body Completion](https://huggingface.co/datasets/YanCORANV/RoboSteer/blob/main/docs/appendix-b3/upper-to-full-body-completion.md) — 4 figures
- [Lower-to-Full Body Completion](https://huggingface.co/datasets/YanCORANV/RoboSteer/blob/main/docs/appendix-b3/lower-to-full-body-completion.md) — 4 figures
- [Target Reaching](https://huggingface.co/datasets/YanCORANV/RoboSteer/blob/main/docs/appendix-b3/target-reaching.md) — 2 figures
## Level 2

### Constraint steering

- [Order](https://huggingface.co/datasets/YanCORANV/RoboSteer/blob/main/docs/appendix-b3/order.md) — 3 figures
- [Times](https://huggingface.co/datasets/YanCORANV/RoboSteer/blob/main/docs/appendix-b3/times.md) — 2 figures
- [Trajectory](https://huggingface.co/datasets/YanCORANV/RoboSteer/blob/main/docs/appendix-b3/trajectory.md) — 2 figures
- [Amplitude](https://huggingface.co/datasets/YanCORANV/RoboSteer/blob/main/docs/appendix-b3/amplitude.md) — 2 figures
- [Body Restrain](https://huggingface.co/datasets/YanCORANV/RoboSteer/blob/main/docs/appendix-b3/body-restrain.md) — 2 figures
- [Direction](https://huggingface.co/datasets/YanCORANV/RoboSteer/blob/main/docs/appendix-b3/direction.md) — 2 figures
- [Speed](https://huggingface.co/datasets/YanCORANV/RoboSteer/blob/main/docs/appendix-b3/speed.md) — 2 figures
## Level 3

### Compositional steering

- [Interleaved Multi-Source Steering](https://huggingface.co/datasets/YanCORANV/RoboSteer/blob/main/docs/appendix-b3/interleaved-multi-source-steering.md) — 1 figure