You can deploy this model and start earning money today!

Cloth-splatters/
dexgarmentlab-folding-lifting-state-est-gps-flow

Name: dexgarmentlab-folding-lifting-state-est-gps-flow
Availability: InStock

diffusers

safetensors

Views

No views yet

dexgarmentlab-folding-lifting-state-est-gps-flow

GPSStateEstModel — graph-based (GNN + Transformer) state estimation for variable-vertex cloth meshes. Reconstructs the full cloth mesh state from a partial point cloud observation via flow matching, conditioned on each cloth's own rest state and topology (no global template).

Task data: DexGarmentLab mixed-garment fold + lift-place demos, point-cloud-cleaned (dexgarmentlab_folding_lifting_meshes_clean.h5, Cloth-splatters/dexgarmentlab-folding-lifting-meshes-clean)
Formulation: flow matching
Rest-position centering: self
Cross-attention mode: parallel
Max vertices per mesh: 2048
Best validation loss: 0.00019553977641302415 (checkpoint in model/ is checkpoint-best)
Training run: dexgarment_flow_clean_2026-08-05_14-28-22_475354 (full config in config.yml)

Retrained on cleaned point clouds (2026-08-06)

This revision supersedes the 2026-08-04 upload, which was trained on dexgarmentlab_folding_lifting_meshes.h5 before the point-cloud audit. Roughly half of those observations carried scene geometry (floor / fixture returns) mixed into the cloth points; the cleaned dataset removes it and trims the affected frames. The previous revision is still reachable by commit hash.

Validation losses are not comparable across that change — the split itself changed, so the 0.000114 recorded on the pre-clean revision does not sit on the same scale as the number above, and is not evidence that the older checkpoint is better.

Rest-position centering: `self`

This checkpoint was trained with the self rest-position centering convention: the per-cloth rest (template) positions fed to the model are centered by subtracting the rest mesh's own (valid-vertex) mean, independent of the observed point cloud.

Implications:

The model is translation-invariant in rest space, which makes it suitable for real-world inference: the canonical template can live in any coordinate frame relative to the camera point cloud.
At inference, the rest positions must be self-centered the same way. Pipelines in the training repo read model.config.rest_pos_centering (recorded as "self" in this checkpoint's config.json) and do this automatically.
Not interchangeable with checkpoints trained with the legacy pcd convention (all state-estimation checkpoints in this org trained on or before 2026-05-10).

v2 (2026-08-20) — this revision

Retrained with the v2 recipe: EMA weights (shipped in model/), heavy occlusion/outlier augmentation (fold-unfold-lift set, scale 0.8-1.25, no yaw) and coord_scale 3. Best validation MSE 2.81e-4 m² (previous release: 8.55e-4); held-out first-frame reconstruction better on 7/8 test cloths, and the TNLC_Jacket001_0 hard start improves from rank 136 to rank 3 in the 239-mesh identification probe. Sequential cross-attention, 200k steps. The previous release remains available at revision 500dc4fe0f492b833fbd5bdeb0a4b852f67e80e0.

v2-cs3-tf2-mismatch (2026-08-31) — this revision

The flow-matching v2 recipe (EMA + heavy augmentation + coord_scale 3) plus two-frame temporal training (num_context_frames: 2, single-frame at inference via the pcd_prev=None fallback) plus template-size mismatch augmentation (template_mismatch_range: [0.5, 2.0]). The combination beats the previous revision on every measured axis: best val 2.67e-4 m² (previous: 2.81e-4), flat across template scale factors 0.5–2 while also better at true size (masked MSE 9.6e-4 vs 1.11e-3 at 5 denoising steps; the previous revision degrades to 2.3e-2 at factor 1.88), real take-2 pants tracked at metric scale with no hand-tuned rescale at 9.66 mm symmetric Chamfer, and the TNLC_Jacket001_0 fold cold start kept at identification rank 5 of 239 (mismatch augmentation without the temporal regularizer loses it to rank 53). Previous revision (flow_v2_cs3): 2644616bf090eb75937624a7e7a26582f1c8cb7c.

Cloth-splatters/dexgarmentlab-folding-lifting-state-est-gps-flow