KMM: Key Frame Mask Mamba for Extended Motion Generation

TL;DR

This paper introduces KMM, a novel architecture that enhances long and complex human motion generation by focusing on key frames, improving multimodal fusion, and achieving state-of-the-art results on the BABEL dataset.

Contribution

The paper proposes KMM with key frame masking, a contrastive learning paradigm for better multimodal fusion, and demonstrates superior performance on human motion generation tasks.

Findings

Achieved over 57% reduction in FID score.

Reduced model parameters by 70% compared to previous methods.

Enhanced focus on key actions in motion segments.

Abstract

Human motion generation is a cut-edge area of research in generative computer vision, with promising applications in video creation, game development, and robotic manipulation. The recent Mamba architecture shows promising results in efficiently modeling long and complex sequences, yet two significant challenges remain: Firstly, directly applying Mamba to extended motion generation is ineffective, as the limited capacity of the implicit memory leads to memory decay. Secondly, Mamba struggles with multimodal fusion compared to Transformers, and lack alignment with textual queries, often confusing directions (left or right) or omitting parts of longer text queries. To address these challenges, our paper presents three key contributions: Firstly, we introduce KMM, a novel architecture featuring Key frame Masking Modeling, designed to enhance Mamba's focus on key actions in motion segments. This approach addresses the memory decay problem and represents a pioneering method in customizing strategic frame-level masking in SSMs. Additionally, we designed a contrastive learning paradigm for addressing the multimodal fusion problem in Mamba and improving the motion-text alignment. Finally, we conducted extensive experiments on the go-to dataset, BABEL, achieving state-of-the-art performance with a reduction of more than 57% in FID and 70% parameters compared to previous state-of-the-art methods. See project website: https://steve-zeyu-zhang.github.io/KMM