4D/3D Vision and Fast Video Diffusion Breakthroughs: Gen4P Single-Pass 4D Perceiver, SUPRA-TEX 32.08 dB PBR Texturing, and HKUST ForcingBooster (-50.19% Patch NFE)

Background and the Problem

3D/4D spatial intelligence and real-time interactive video diffusion face three pressing bottlenecks: first, repurposing pretrained video diffusion models for 4D reconstruction and 3D point tracking typically relies on slow multi-step denoising; second, 3D texture generators (TRELLIS.2, Hunyuan3D 2.1) struggle with multi-view seams, baked-in lighting artifacts, and texture-geometry misalignment; third, autoregressive video diffusion models (Causal Forcing, Wan2.1) apply uniform 4-step denoising across every spatial patch, wasting compute on static backgrounds. Gen4P (arXiv:2610.22371), HUST and Huawei's SUPRA-TEX (arXiv:2610.22352), and HKUST's ForcingBooster (arXiv:2610.12625) tackle each of these bottlenecks directly.

Architecture and How It Works

1. Gen4P: Single-Pass Feed-Forward Continuous-Time 4D Perceiver (arXiv:2610.22371):

• Converts a pretrained video generator into a single-pass feed-forward 4D perception model without iterative denoising. It decouples spatial and temporal aggregation via continuous-time on-demand queries and injects a frozen video-depth LoRA to surface latent geometric priors for simultaneous 4D point cloud reconstruction and 3D point tracking.

1. SUPRA-TEX: Read-Only Native Latent Anchors & Interleaved Flow Transformer (arXiv:2610.22352):

• Decouples illumination removal and unseen surface synthesis: a fine-tuned MLLM first estimates de-lit PBR maps directly from the reference image; visible surface attributes are projected into the native 3D VAE latent space as read-only geometric anchors; and an Interleaved Flow Transformer completes occluded regions by combining global image attention with local 3D structural attention.

1. ForcingBooster: Adaptive Patch-Wise Denoising for Autoregressive Video Diffusion (arXiv:2610.12625):

• Dynamically allocates denoising steps per spatial patch in autoregressive video diffusion so static or low-motion patches exit early while complex dynamic patches receive full multi-step refinement.

Benchmarks and Measured Results

• Gen4P (WorldTrack 3D Point Tracking Benchmark): Achieves SOTA APD ↑ / EPE ↓ of 91.93 / 0.112 on ADT, 88.18 / 0.130 (92.52 / 0.090 dynamic) on Dynamic Replica, and 82.11 / 0.149 on Panoptic Studio, outperforming MonST3R, STv2, St4RTrack, Any4D, and OpenD4RT.
• SUPRA-TEX (DTC-Tex 3D PBR Benchmark): Reaches 32.08 dB Shaded PBR PSNR, 23.67 dB foreground PSNR, 0.042 LPIPS, and 16.31 FD_DINOv3, decisively beating Pixal3D (29.19 dB), TRELLIS.2 (28.60 dB), LumiTex (28.29 dB), and Hunyuan3D 2.1 (26.80 dB).
• ForcingBooster (VBench Autoregressive Video Acceleration): Cuts patch denoising NFE by 50.19% and reduces end-to-end latency (including VAE decoding) from 11.15s to 8.56s (1.30x speedup) while preserving the Causal Forcing Semantic score (72.20 vs. 72.21 4-step baseline), verified across Wan2.1 14B, LingBot-World-V2, and LongLive.

Getting Started for Developers

All three projects provide open-source repositories (yitongx/Gen4P, Supra-3D/SUPRA-Tex, and IGL-HKUST/ForcingBooster):

• Teams serving interactive world models or streaming video generators (Wan2.1 14B, LongLive, LingBot-World-V2) can plug in ForcingBooster's patch-wise early exit schedule to halve patch denoising FLOPs without degrading visual semantics;
• 3D asset pipelines can adopt SUPRA-TEX to preserve pixel-exact frontal textures while synthesizing relightable PBR materials for occluded regions.