proomt

Search

Search posts, papers, and topics

All posts

Hugging Face Daily PapersYang-Tian Sun, Tianjia Liu, Zehuan Huang1 min readpaperadvanced

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

Read on arxiv.org

Summary

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.

  • CCM provides per‑pixel correspondences to a canonical object coordinate system, enabling robust dense alignment to scene space.
  • A Point Cloud Map (PCM) from monocular geometry supplies scene‑level structure, while CCM supplies object‑level detail.
  • The diffusion transformer uses separate expert streams for geometry and CCM, sharing attention to enforce layout consistency.
  • Training requires only object‑level 3D data; no explicit scene layout labels are needed, improving scalability.

3D scene generation engineers and researchers will care because it offers a scalable way to achieve accurate object placement without costly scene‑level annotations.

8/10
diffusiongraphicscomputer vision3d reconstructiongenerative models

Related reading

  3. FLAT: Resampling Image and Text into 1D Flexible-Length Aligned Transmodal Tokens for Retrieval and Generation

    FLAT is a pre‑training framework that encodes images and text into a shared 1D token sequence with variable length via nested dropout, enabling the same embeddings for cross‑modal retrieval and generation. It attains state‑of‑the‑art scores on COCO and Flickr30K for captioning, retrieval, and text‑to‑image generation, and supports interpolation and zero‑shot composed retrieval.

    Hugging Face Daily Papersarxiv.org1 minpaper

  4. UFO: Chain-of-Evaluation for Omni-Condition Alignment in Multi-Modal Image Generation

    UFO introduces an Atomized Chain‑of‑Evaluation (AEU) framework that breaks omni‑condition alignment in multi‑modal image generation into a sequential set of fine‑grained checks, achieving a 15.25 % boost in correlation with human judgments. The authors also release UFO‑Bench, a benchmark for testing how well models satisfy combined textual and visual conditions.

    Hugging Face Daily Papersarxiv.org1 minpaper

  5. SNAP3D: Physically Grounded 3D Parts for Assembly from a Single Image

    The paper presents SNAP3D, a physics‑guided pipeline that converts a single image into a set of 3D parts that can be assembled without interpenetration. By using simulation‑driven connector placement and a new physics‑based evaluation, the method yields assemblies that are both geometrically accurate and stable enough for 3D printing.

    Hugging Face Daily Papersarxiv.org1 minpaper

  6. Breaking the Vision-Action Shortcut: Latent Interface Training for Generalizable Robotics Foundation Models

    Latent Interface Training (LIT) first learns a goal‑conditioned action prior without visual input, then adds a pose‑supervised latent interface as the only visual conditioning path. Applied to several vision‑language‑action models, LIT cuts vision‑action shortcuts and lifts LIBERO‑Plus success by 3.9–10.7 points and real‑world task success by 13.3–16.7 points under distribution shifts.

    Hugging Face Daily Papersarxiv.org1 minpaper