Signals 4 · free daily AI digest

Kaininja: Extending Native 3D Generators to the Part Level

Paper recorded by Signals 4 on 2026-09-14 in cs.CV. Abstract reproduced from arXiv; link to the original below.

Published 2026-09-14 on arXiv · recorded by Signals 4 on 2026-09-15

Category: cs.CV · 计算机视觉 · first seen 2026-09-15

Abstract

Native 3D generators turn one image into a single mesh. TRELLIS.2 and its peers deliver high-fidelity non-watertight geometry with materials, but the output is one fused object, while downstream work such as editing, rigging and simulation operates on part-level assets. A naive idea is to run a 3D segmentation network on the fused mesh that TRELLIS.2 generates, but such pipelines are slow and boun

Read on arXiv →

#60 most recent of 237 cs.CV papers we have recorded · ↑ newer: Circuit-MLLM: Topological Logic-Guided Latent-Space Visual Reasoning f · ↓ older: From Model Patterns to Abstract Semantics in Compositional Zero-Shot L
Cite this page: Kaininja: Extending Native 3D Generators to the Part Level: the #60 most recent of 237 cs.CV papers we have recorded (as of 2026-09-14). Source: Signals 4 (Signals API) — https://data.jiangzhang.ca/signals4/t/papers/kaininja-extending-native-3d-generators-to-the-part-level.html
Free to quote with attribution to “Signals 4 (Signals API)”. Machine-readable: papers.json
Related: More cs.CV papers · arXiv signals · All papers · Today in AI
Get 4 AI signals a day by email — free.
Subscribe free → See all plans →
Get 4 AI signals a day by email — free
All models · All repos · By company · Daily editions