English static mirror for SEO/GEO · AI-assisted translation · Read Chinese original

LottieGPT: Tokenizing Vector Animation for Autoregressive Generation

Forum topic · 小凯 · 2026-04-15

Summary

LottieGPT (arXiv:2604.11792) introduces the first framework for tokenizing and autoregressively generating vector animation. While video generation has advanced rapidly, existing generative models operate exclusively in raster space and cannot produce vector animations, which offer resolution-independence, compactness, semantic structure, and editable parametric motion. The authors build on Lottie, a widely deployed JSON animation standard, and design a custom Lottie Tokenizer that encodes hierarchical geometric primitives, transforms, and keyframe-based motion into compact, semantically aligned token sequences. They construct LottieAnimation-660K, a dataset of 660K vector animations, and fine-tune Qwen-VL to create LottieGPT, a model capable of generating coherent, editable vector animations directly from natural language or visual prompts. The work, from researchers including Junhao Chen, Kejun Gao, and Ruqi Huang, bridges generative AI with structured, editable multimedia content.

[Paper] LottieGPT: Tokenizing Vector Animation for Autoregressive Generation

Paper Overview

  • Research area: cs.CV
  • Authors: Junhao Chen, Kejun Gao, Yuehan Cui, Mingze Sun, Mingjin Chen, Shaohui Wang, Xiaoxiao Long, Fei Ma, Qi Tian, Ruqi Huang, Hao Zhao
  • Published: 2026-04-13
  • arXiv: 2604.11792
  • Summary

    Despite rapid progress in video generation, existing models are incapable of producing vector animation, a dominant and highly expressive form of multimedia on the Internet. Vector animations offer resolution-independence, compactness, semantic structure, and editable parametric motion representations, yet current generative models operate exclusively in raster space and thus cannot synthesize them.

    This paper presents the first framework for tokenizing and autoregressively generating vector animation. Building on Lottie, a widely deployed JSON-based animation standard, the authors design a custom Lottie Tokenizer that encodes hierarchical geometric primitives, transforms, and keyframe-based motion into compact, semantically aligned token sequences. They further construct the LottieAnimation-660K dataset and fine-tune Qwen-VL to create LottieGPT, a model capable of generating coherent, editable vector animations directly from natural language or visual prompts.

    Key Contributions

  • Lottie Tokenizer: encodes hierarchical geometry, transforms, and keyframe motion as compact, semantic token sequences
  • LottieAnimation-660K: a large-scale dataset for vector animation generation
  • LottieGPT: a Qwen-VL-based autoregressive model producing editable vector animations from text or visual prompts
--- *Auto-collected on 2026-04-15*

Tags

#lottiegpt#vector-animation#autoregressive-generation#qwen-vl#tokenizer#generative-models#computer-vision#arxiv

This page is an English static mirror generated for search and AI citation. It may be a full translation or structured summary of the Chinese original. Canonical interactive discussion lives on the Chinese page: https://zhichai.net/topic/177618479