CIA: Controllable Image Augmentation Framework Based on Stable Diffusion

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Benkedadra, Mohamed, Rimez, Dany, Godelaine, Tiffanie, Chidambaram, Natarajan, Khosroshahi, Hamed Razavi, Tellez, Horacio, Mancas, Matei, Macq, Benoit, Mahmoudi, Sidi Ahmed
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910712979259392
author Benkedadra, Mohamed
Rimez, Dany
Godelaine, Tiffanie
Chidambaram, Natarajan
Khosroshahi, Hamed Razavi
Tellez, Horacio
Mancas, Matei
Macq, Benoit
Mahmoudi, Sidi Ahmed
author_facet Benkedadra, Mohamed
Rimez, Dany
Godelaine, Tiffanie
Chidambaram, Natarajan
Khosroshahi, Hamed Razavi
Tellez, Horacio
Mancas, Matei
Macq, Benoit
Mahmoudi, Sidi Ahmed
contents Computer vision tasks such as object detection and segmentation rely on the availability of extensive, accurately annotated datasets. In this work, We present CIA, a modular pipeline, for (1) generating synthetic images for dataset augmentation using Stable Diffusion, (2) filtering out low quality samples using defined quality metrics, (3) forcing the existence of specific patterns in generated images using accurate prompting and ControlNet. In order to show how CIA can be used to search for an optimal augmentation pipeline of training data, we study human object detection in a data constrained scenario, using YOLOv8n on COCO and Flickr30k datasets. We have recorded significant improvement using CIA-generated images, approaching the performances obtained when doubling the amount of real images in the dataset. Our findings suggest that our modular framework can significantly enhance object detection systems, and make it possible for future research to be done on data-constrained scenarios. The framework is available at: github.com/multitel-ai/CIA.
format Preprint
id arxiv_https___arxiv_org_abs_2411_16128
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle CIA: Controllable Image Augmentation Framework Based on Stable Diffusion
Benkedadra, Mohamed
Rimez, Dany
Godelaine, Tiffanie
Chidambaram, Natarajan
Khosroshahi, Hamed Razavi
Tellez, Horacio
Mancas, Matei
Macq, Benoit
Mahmoudi, Sidi Ahmed
Computer Vision and Pattern Recognition
Artificial Intelligence
Computer vision tasks such as object detection and segmentation rely on the availability of extensive, accurately annotated datasets. In this work, We present CIA, a modular pipeline, for (1) generating synthetic images for dataset augmentation using Stable Diffusion, (2) filtering out low quality samples using defined quality metrics, (3) forcing the existence of specific patterns in generated images using accurate prompting and ControlNet. In order to show how CIA can be used to search for an optimal augmentation pipeline of training data, we study human object detection in a data constrained scenario, using YOLOv8n on COCO and Flickr30k datasets. We have recorded significant improvement using CIA-generated images, approaching the performances obtained when doubling the amount of real images in the dataset. Our findings suggest that our modular framework can significantly enhance object detection systems, and make it possible for future research to be done on data-constrained scenarios. The framework is available at: github.com/multitel-ai/CIA.
title CIA: Controllable Image Augmentation Framework Based on Stable Diffusion
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2411.16128