MoVer: Motion Verification for Motion Graphics Animations
Open Access
- 27 July 2025
- journal article
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Graphics
- Vol. 44 (4) , 1-17
Abstract
While large vision-language models can generate motion graphics animations from text prompts, they regularly fail to include all spatio-temporal properties described in the prompt. We introduce MoVer, a motion verification DSL based on first-order logic that can check spatio-temporal properties of a motion graphics animation. We identify a general set of such properties that people commonly use to describe animations (e.g., the direction and timing of motions, the relative positioning of objects, etc.). We implement these properties as predicates in MoVer and provide an execution engine that can apply a MoVer program to any input SVG-based motion graphics animation. We then demonstrate how MoVer can be used in an LLM-based synthesis and verification pipeline for iteratively refining motion graphics animations. Given a text prompt, our pipeline synthesizes a motion graphics animation and a corresponding MoVer program. Executing the verification program on the animation yields a report of the predicates that failed and the report can be automatically fed back to LLM to iteratively correct the animation. To evaluate our pipeline, we build a synthetic dataset of 5600 text prompts paired with ground truth MoVer verification programs. We find that while our LLM-based pipeline is able to automatically generate a correct motion graphics animation for 58.8% of the test prompts without any iteration, this number raises to 93.6% with up to 50 correction iterations. Our code and dataset are at https://mover-dsl.github.io.Keywords
All Related Versions
Funding Information
- National Science Foundation (2219864)
This publication has 26 references indexed in Scilit:
- SparseCtrl: Adding Sparse Controls to Text-to-Video Diffusion Models2024
- VideoCrafter2: Overcoming Data Limitations for High-Quality Video Diffusion Models2024
- Breathing Life Into Sketches Using Text-to-Video Priors2024
- Predicated Diffusion: Predicate Logic-Based Attention Guidance for Text-to-Image Diffusion Models2024
- ViperGPT: Visual Inference via Python Execution for Reasoning2023
- A Review of Text-to-Animation SystemsIEEE Access, 2023
- Visual Programming: Compositional visual reasoning without training2023
- Spatial reasoning with rectangular cardinal relationsAnnals of Mathematics and Artificial Intelligence, 2013
- DECOMPOSING A MATRIX INTO SIMPLE TRANSFORMATIONS1991
- Maintaining knowledge about temporal intervalsCommunications of the ACM, 1983