80991a8fe9
Every SparseConv3d in the Pixal3D model is (channels, out, 3) - stride=1/padding=None, so spconv dispatches to SubMConv3d, which trellis_sparse_mlx already implements and benchmarks. attn_mode is 'full' throughout. The container split (VarLenTensor / SparseTensor) and the get/register_spatial_cache spelling are already in the shared core. Remaining gap is two ops: SparseUpsample and SparseSpatial2Channel, both cache-paired with a matching downsample rather than recomputing structure.
57 lines
2.3 KiB
Markdown
57 lines
2.3 KiB
Markdown
# pixal3d_mrp_mlx
|
|
|
|
MLX port of [Pixal3D](https://github.com/TencentARC/Pixal3D) (TencentARC + Tsinghua,
|
|
SIGGRAPH 2026, MIT) — pixel-aligned single-image 3D generation — for Apple Silicon.
|
|
|
|
Built on **[`trellis_sparse_mlx`](../trellis_sparse_mlx)**, the shared TRELLIS-lineage
|
|
sparse core. Pixal3D and LATO.2 inherit the same sparse module from TRELLIS.2, so the
|
|
expensive part — submanifold sparse convolution, which has no Metal implementation — is
|
|
already done and tested there.
|
|
|
|
## Why Pixal3D
|
|
|
|
It back-projects pixel features directly into 3D rather than injecting them through
|
|
attention, so silhouettes stay exact to the source image. Different failure mode from
|
|
TRELLIS/Hunyuan, and complementary to them.
|
|
|
|
Upstream needs ~24 GB VRAM, which is a wall on consumer Nvidia and a non-issue on a
|
|
128 GB+ Ultra.
|
|
|
|
## Scope, measured against the shared core
|
|
|
|
| Need | Status |
|
|
|---|---|
|
|
| `SparseConv3d` — **every call is `(c, out, 3)`**, i.e. stride=1/padding=None → SubMConv3d | ✅ in shared core |
|
|
| `attn_mode='full'` (the only mode used) | ✅ in shared core |
|
|
| `SparseLinear`, norms, activations, ResBlock, transformer blocks | ✅ in shared core |
|
|
| `SparseDownsample(2)` | ✅ in shared core |
|
|
| `VarLenTensor` / `SparseTensor` split + `get/register_spatial_cache` | ✅ added to shared core |
|
|
| dense `nn.Conv3d(.., 2, stride=2)` in `sparse_structure_vae` | ✅ maps to `mlx.nn.Conv3d` |
|
|
| `SparseUpsample(2)` | ❌ **to do** — cache-paired inverse of a downsample |
|
|
| `SparseSpatial2Channel(2)` | ❌ **to do** — sparse pixel-shuffle, spatial→channel |
|
|
|
|
So the gap is **two ops**, both of which work through the spatial cache (they pair with a
|
|
matching downsample rather than recomputing structure). Everything else is already
|
|
covered by work done for LATO.2.
|
|
|
|
## Model surface
|
|
|
|
```
|
|
pixal3d/models/
|
|
sparse_structure_vae.py dense Conv3d — voxel structure
|
|
sparse_structure_flow.py structure flow (SS)
|
|
structured_latent_flow.py SLAT flow
|
|
sc_vaes/sparse_unet_vae.py the only file using sparse conv
|
|
```
|
|
|
|
Weights: 24.04 GB across 19 files (1.3B DiTs at 512/1024 + shape/tex decoders).
|
|
|
|
## Status
|
|
|
|
- [x] Scoped against the shared core — gap is two ops
|
|
- [x] Weights downloaded
|
|
- [ ] `SparseUpsample`, `SparseSpatial2Channel`
|
|
- [ ] Weight converter
|
|
- [ ] Model graphs
|
|
- [ ] End-to-end
|