跳到正文
HuggingFace Daily Papers(社區熱門論文)·· 2026-10-20AI 評分45

SolarRenderer:四步生成式參考影片渲染

SolarRenderer: Few-Step Generative Reference Rendering

AI 導讀

SolarRenderer 以凍結的 MiniMax-H3 33B 影片骨幹網絡為基礎,加入輕量級 LoRA,將多種結構輸入渲染成寫實影片。模型以四步生成,速度比 50 步取樣快 12.3 倍;串流版本可即時渲染 768p 影片。訓練數據集包含 136K 段影片,測試顯示模型可泛化至 11 種未見過的結構表示。

正文

Abstract

SolarRenderer adapts a frozen video backbone with lightweight LoRAs to render realistic videos from diverse structural inputs in four steps or real-time streaming, using frame-aligned training data.

Game engines and coding agents can specify the layout, motion, and camera of a scene, yet turning such coarse structure into realistic video requires a renderer that follows the structure and completes the appearance it leaves open. Existing renderers are tied to the representation of one engine, need tens of denoising steps or a costly distillation of their own, and learn from conditions that are not aligned with their target videos. We present SolarRenderer, a structure-conditioned video renderer that adapts a frozen MiniMax-H3 33B backbone with two lightweight LoRAs. Structural videos enter the native reference pathway of the backbone at the positions of the frames they describe, without a type identifier and described only in plain text, so that a single control LoRA follows arbitrary structural representations. Starting from a pretrained four-step LoRA, structure-guided distribution matching renders in four steps, 12.3 times faster than 50-step sampling, while preserving the learned control. A block-causal variant extends the same design to streaming and renders 768p video in real time. To train the model, we build a frame-aligned corpus of 136K clips from game engines, simulators, and YouTube videos. Experiments on an in-distribution test set and a generalization benchmark of 16 structural representations show that SolarRenderer achieves state-of-the-art results across both, and generalizes to 11 unseen representations, including pose skeletons, drawings, and point clouds.

View arXiv page View PDF Project page GitHub 153 Add to collection

Get this paper in your agent:

hf papers read 2610.28630

Don't have the latest CLI?

curl -LsSf https://hf.co/cli/install.sh | bash

Models citing this paper 1

Enderfga/SolarRenderer

Video-to-Video •

Updated 25 minutes ago

•

4

Datasets citing this paper 1

Enderfga/SolarRenderer-Data

Preview •

Updated 25 minutes ago

•

4.46k

•

9

Spaces citing this paper 0

No Space linking this paper

Cite arxiv.org/abs/2610.28630 in a Space README.md to link it from this page.

Collections including this paper 1

來源:HuggingFace Daily Papers(社區熱門論文) · huggingface.co