Saved in:
Bibliographic Details
Main Authors: Radouane, Karim, Lagarde, Julien, Ranwez, Sylvie, Tchechmedjiev, Andon
Format: Preprint
Published: 2023
Subjects:
Online Access:https://arxiv.org/abs/2310.07324
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929482893361152
author Radouane, Karim
Lagarde, Julien
Ranwez, Sylvie
Tchechmedjiev, Andon
author_facet Radouane, Karim
Lagarde, Julien
Ranwez, Sylvie
Tchechmedjiev, Andon
contents Diverse and extensive work has recently been conducted on text-conditioned human motion generation. However, progress in the reverse direction, motion captioning, has seen less comparable advancement. In this paper, we introduce a novel architecture design that enhances text generation quality by emphasizing interpretability through spatio-temporal and adaptive attention mechanisms. To encourage human-like reasoning, we propose methods for guiding attention during training, emphasizing relevant skeleton areas over time and distinguishing motion-related words. We discuss and quantify our model's interpretability using relevant histograms and density distributions. Furthermore, we leverage interpretability to derive fine-grained information about human motion, including action localization, body part identification, and the distinction of motion-related words. Finally, we discuss the transferability of our approaches to other tasks. Our experiments demonstrate that attention guidance leads to interpretable captioning while enhancing performance compared to higher parameter-count, non-interpretable state-of-the-art systems. The code is available at: https://github.com/rd20karim/M2T-Interpretable.
format Preprint
id arxiv_https___arxiv_org_abs_2310_07324
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Guided Attention for Interpretable Motion Captioning
Radouane, Karim
Lagarde, Julien
Ranwez, Sylvie
Tchechmedjiev, Andon
Computer Vision and Pattern Recognition
Diverse and extensive work has recently been conducted on text-conditioned human motion generation. However, progress in the reverse direction, motion captioning, has seen less comparable advancement. In this paper, we introduce a novel architecture design that enhances text generation quality by emphasizing interpretability through spatio-temporal and adaptive attention mechanisms. To encourage human-like reasoning, we propose methods for guiding attention during training, emphasizing relevant skeleton areas over time and distinguishing motion-related words. We discuss and quantify our model's interpretability using relevant histograms and density distributions. Furthermore, we leverage interpretability to derive fine-grained information about human motion, including action localization, body part identification, and the distinction of motion-related words. Finally, we discuss the transferability of our approaches to other tasks. Our experiments demonstrate that attention guidance leads to interpretable captioning while enhancing performance compared to higher parameter-count, non-interpretable state-of-the-art systems. The code is available at: https://github.com/rd20karim/M2T-Interpretable.
title Guided Attention for Interpretable Motion Captioning
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2310.07324