| license: apache-2.0 | |
| tags: [gaussian-splatting, 3d-reconstruction, novel-view-synthesis, feed-forward] | |
| datasets: [DL3DV/DL3DV-10K] | |
| # ReTokenGS — grad | |
| Recurrent Gaussian Tokens for query-based feed-forward 3D Gaussian Splatting, built on [LocusGS](https://github.com/leo-frank/LocusGS) / [TokenGS](https://github.com/nv-tlabs/TokenGS): the 12 independent decoder layers are replaced by **one weight-shared layer unrolled 12 times**, and after every round each token reads the **gradient of the input-view rendering loss with respect to its own Gaussians** (after G3R); no error cross-attention. 4 posed input views, 256×256, 1024 tokens × 64 Gaussians. | |
| `model.safetensors` is the average of the last six training checkpoints (epochs 44–66). It also contains the weights of decoder blocks 1–11 inherited from initialization; only block 0 is executed. | |
| | Model (official DL3DV benchmark: 140 scenes, 4 views, 256×256) | Tokens | PSNR | LPIPS | SSIM | | |
| |---|---|---|---|---| | |
| | LocusGS stage 1, released `base` weights (same evaluation) | 1024 | 24.154 | 0.2766 | 0.7805 | | |
| | **ReTokenGS-grad (this file)** | 1024 | **24.733** | **0.2664** | **0.7916** | | |
| | ReTokenGS-dens, 4096 tokens on demand ([`Yusin/ReTokenGS-dens`](https://huggingface.co/Yusin/ReTokenGS-dens)) | ≤ 4096 | 25.091 | 0.2442 | 0.8031 | | |
| Code, evaluation and training commands: GitHub [`Yusin2Chen/TokenGS-Loop`, branch `grad`](https://github.com/Yusin2Chen/TokenGS-Loop/tree/grad). License: Apache-2.0 (derived from LocusGS / TokenGS). | |
| The numbers were produced with the released code, which has no error cross-attention; the `loop_err_*` tensors in this file are unused and ignored at load time. | |