proomt

Search

Search posts, papers, and topics

All posts

Hugging Face Daily PapersYibo Zhang, Ze Yuan, Nan Cao1 min readpaperadvanced

UltraTex: Unleashing 2K Multi-View Diffusion for 3D Texturing

Read on arxiv.org

Summary

UltraTex proposes an efficient pipeline for 2K multi‑view diffusion‑based 3D texturing by dropping background tokens, using block‑sparse attention, and a foreground‑aware VAE decoder. The method yields 20‑91× training and 22‑75× inference speedups while preserving fine‑grained texture detail.

  • Background Token Dropping removes background pixels before the DiT backbone, cutting token count dramatically.
  • Block‑Sparse Attention limits attention computation to foreground tokens, further reducing memory and latency.
  • Foreground‑Aware VAE Decoding reconstructs high‑resolution views without artifacts despite foreground‑only processing.
  • TexVerse dataset provides 268k ultra‑high‑resolution multi‑view renders to train 2K diffusion models.

Graphics engineers and diffusion researchers should care because it makes high‑resolution 3D texturing practical at scale.

8/10
diffusioncomputer visionml infragpu3d texturing

Related reading

  2. Anatomy of a Texture

    A practical deep‑dive into modern texture memory layout: block‑compression (BC7), swizzle ordering (Morton/Z‑order), mip‑map hierarchies, and concrete debugging tricks for cross‑platform texture conversion.

    Hacker News front pagegithub.io12 minHN10816

  3. Video DeltaNet: A Video-Native Hybrid Attention for Livestream Video Generation

    Video DeltaNet (VDN) replaces full‑softmax attention in video diffusion models with a hybrid: per‑frame local Softmax for fine detail and a bidirectional linear memory (Video Delta Attention) for long‑range context. A teacher‑alignment schedule injects the linear branch into a pretrained MiniMax H3 model, preserving Softmax for text/audio streams. On eight NVIDIA B200 GPUs VDN‑H3 denoises a 14.3‑…

    Hugging Face Daily Papersarxiv.org1 minpaper

  4. Mira-Scene: Pixel-Aligned Layouts for Generative 3D Scene

    Mira-Scene proposes a compositional 3D scene reconstruction pipeline that replaces sparse pose regression with a dense Canonical Coordinate Map (CCM) linking image pixels to bounded object space. Coupled with a diffusion transformer, it achieves up to 40% higher 3D‑IoU than prior methods without scene‑level layout annotations.

    Hugging Face Daily Papersarxiv.org1 minpaper

  5. VC-Attention: Value Smoothing and Softmax Casting for Low-bit Attention

    VC-Attention introduces a training‑free low‑bit attention pipeline for diffusion transformers. It smooths value tensors via lightweight online clustering (V‑Smooth) and quantizes only the residual after subtracting block means, restoring the mean from the softmax row sum. It also replaces the FP32 softmax exponential with a fused FP8 cast (ExpCast‑FP8) that maps log‑scores directly to E4M3 probab…

    Hugging Face Daily Papersarxiv.org1 minpaper

  6. Why Do Video Diffusion Models Violate Physics? Unveiling the Flaws in Attention Mechanisms

    The paper investigates why video diffusion models often break physical laws, pinpointing excessive spatial decay from Rotary Position Embedding (RoPE) as the culprit. By analyzing cross‑attention trajectories and self‑attention patterns, the authors identify specific attention heads that drive motion planning. They propose a lightweight fix: scaling RoPE frequency per denoising step, which empiri…

    Hugging Face Daily Papersarxiv.org1 minpaper