MotionBank: A Large-scale Video Motion Benchmark with Disentangled Rule-based Annotations

TL;DR

MotionBank introduces a large-scale, diverse video motion dataset with rule-based annotations to improve the development and benchmarking of versatile motion models for various human motion tasks.

Contribution

The paper presents MotionBank, a comprehensive dataset with 13 datasets, 1.24 million motion sequences, and rule-based text annotations, addressing limitations of previous small-scale and context-limited motion datasets.

Findings

MotionBank enhances human motion generation and understanding tasks.

Rule-based annotations improve motion-text alignment.

Large-scale dataset benefits general motion modeling.

Abstract

In this paper, we tackle the problem of how to build and benchmark a large motion model (LMM). The ultimate goal of LMM is to serve as a foundation model for versatile motion-related tasks, e.g., human motion generation, with interpretability and generalizability. Though advanced, recent LMM-related works are still limited by small-scale motion data and costly text descriptions. Besides, previous motion benchmarks primarily focus on pure body movements, neglecting the ubiquitous motions in context, i.e., humans interacting with humans, objects, and scenes. To address these limitations, we consolidate large-scale video action datasets as knowledge banks to build MotionBank, which comprises 13 video action datasets, 1.24M motion sequences, and 132.9M frames of natural and diverse human motions. Different from laboratory-captured motions, in-the-wild human-centric videos contain abundant motions in context. To facilitate better motion text alignment, we also meticulously devise a motion caption generation algorithm to automatically produce rule-based, unbiased, and disentangled text descriptions via the kinematic characteristics for each motion. Extensive experiments show that our MotionBank is beneficial for general motion-related tasks of human motion generation, motion in-context generation, and motion understanding. Video motions together with the rule-based text annotations could serve as an efficient alternative for larger LMMs. Our dataset, codes, and benchmark will be publicly available at https://github.com/liangxuy/MotionBank.