Depth-Guided Self-Supervised Human Keypoint Detection via Cross-Modal Distillation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Anand, Aman, Rashno, Elyas, Eskandari, Amir, Zulkernine, Farhana
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916894534008832
author Anand, Aman
Rashno, Elyas
Eskandari, Amir
Zulkernine, Farhana
author_facet Anand, Aman
Rashno, Elyas
Eskandari, Amir
Zulkernine, Farhana
contents Existing unsupervised keypoint detection methods apply artificial deformations to images such as masking a significant portion of images and using reconstruction of original image as a learning objective to detect keypoints. However, this approach lacks depth information in the image and often detects keypoints on the background. To address this, we propose Distill-DKP, a novel cross-modal knowledge distillation framework that leverages depth maps and RGB images for keypoint detection in a self-supervised setting. During training, Distill-DKP extracts embedding-level knowledge from a depth-based teacher model to guide an image-based student model with inference restricted to the student. Experiments show that Distill-DKP significantly outperforms previous unsupervised methods by reducing mean L2 error by 47.15% on Human3.6M, mean average error by 5.67% on Taichi, and improving keypoints accuracy by 1.3% on DeepFashion dataset. Detailed ablation studies demonstrate the sensitivity of knowledge distillation across different layers of the network. Project Page: https://23wm13.github.io/distill-dkp/
format Preprint
id arxiv_https___arxiv_org_abs_2410_14700
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Depth-Guided Self-Supervised Human Keypoint Detection via Cross-Modal Distillation
Anand, Aman
Rashno, Elyas
Eskandari, Amir
Zulkernine, Farhana
Computer Vision and Pattern Recognition
Artificial Intelligence
Existing unsupervised keypoint detection methods apply artificial deformations to images such as masking a significant portion of images and using reconstruction of original image as a learning objective to detect keypoints. However, this approach lacks depth information in the image and often detects keypoints on the background. To address this, we propose Distill-DKP, a novel cross-modal knowledge distillation framework that leverages depth maps and RGB images for keypoint detection in a self-supervised setting. During training, Distill-DKP extracts embedding-level knowledge from a depth-based teacher model to guide an image-based student model with inference restricted to the student. Experiments show that Distill-DKP significantly outperforms previous unsupervised methods by reducing mean L2 error by 47.15% on Human3.6M, mean average error by 5.67% on Taichi, and improving keypoints accuracy by 1.3% on DeepFashion dataset. Detailed ablation studies demonstrate the sensitivity of knowledge distillation across different layers of the network. Project Page: https://23wm13.github.io/distill-dkp/
title Depth-Guided Self-Supervised Human Keypoint Detection via Cross-Modal Distillation
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2410.14700