Saved in:
Bibliographic Details
Main Authors: Khurana, Mannat, Jain, Sanyam, Agarwal, Rishav
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2605.27203
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914605110919168
author Khurana, Mannat
Jain, Sanyam
Agarwal, Rishav
author_facet Khurana, Mannat
Jain, Sanyam
Agarwal, Rishav
contents Animation elevates digital documents into immersive experiences, yet creating custom motion paths remains cumbersome, requiring designers to manually select presets, plot Bézier points, and configure timing properties. We introduce Generative Animations, a system that transforms natural language prompts into production-ready animations. By chaining Large Language Models (LLMs) for semantic parsing with the Segment Anything Model (SAM) for visual grounding, our pipeline automatically generates motion paths that respect scene geometry, handle depth-based occlusions, and honor 3D perspective transforms. We demonstrate the system through three use cases: contour-following trajectories, orbital animations with z-order awareness, and perspective-aligned motion on transformed objects.
format Preprint
id arxiv_https___arxiv_org_abs_2605_27203
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Generative Animations: A Multi-Model Pipeline for Prompt-Driven Motion Synthesis
Khurana, Mannat
Jain, Sanyam
Agarwal, Rishav
Computer Vision and Pattern Recognition
Artificial Intelligence
Animation elevates digital documents into immersive experiences, yet creating custom motion paths remains cumbersome, requiring designers to manually select presets, plot Bézier points, and configure timing properties. We introduce Generative Animations, a system that transforms natural language prompts into production-ready animations. By chaining Large Language Models (LLMs) for semantic parsing with the Segment Anything Model (SAM) for visual grounding, our pipeline automatically generates motion paths that respect scene geometry, handle depth-based occlusions, and honor 3D perspective transforms. We demonstrate the system through three use cases: contour-following trajectories, orbital animations with z-order awareness, and perspective-aligned motion on transformed objects.
title Generative Animations: A Multi-Model Pipeline for Prompt-Driven Motion Synthesis
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2605.27203