Efficient Object Placement via FTOPNet
-
Published:2023-09-30
Issue:19
Volume:12
Page:4106
-
ISSN:2079-9292
-
Container-title:Electronics
-
language:en
-
Short-container-title:Electronics
Author:
Ye Guosheng12ORCID, Wang Jianming12ORCID, Yang Zizhong1
Affiliation:
1. Yunnan Provincial Key Laboratory of Entomological Biopharmaceutical R&D, Dali University, Dali 671000, China 2. School of Mathematics and Computer Science, Dali University, Dali 671003, China
Abstract
Image composition involves the placement of foreground objects at an appropriate scale within a background image to create a visually realistic composite image. However, manual operations for this task are time-consuming and labor-intensive. In this study, we propose an efficient method for foreground object placement, comprising a background feature extraction module (BFEM) designed for background images and a foreground–background cross-attention feature fusion module (FBCAFFM). The BFEM is capable of extracting precise and comprehensive information from the background image. The fused features enable the network to learn additional information related to foreground–background matching, aiding in the prediction of foreground object placement and size. Our experiments are conducted using the publicly available object placement assessment (OPA) dataset. Both quantitative and visual results demonstrate that FTOPNet effectively performs the foreground object placement task and offers a practical solution for image composition tasks.
Funder
National Natural Science Foundation of China Yunnan Fundamental Research Projects
Subject
Electrical and Electronic Engineering,Computer Networks and Communications,Hardware and Architecture,Signal Processing,Control and Systems Engineering
Reference34 articles.
1. Niu, L., Cong, W., Liu, L., Hong, Y., Zhang, B., Liang, J., and Zhang, L. (2022). Making Images Real Again: A Comprehensive Survey on Deep Image Composition. arXiv. 2. Chen, B.C., and Kae, A. (2019, January 15–20). Toward realistic image compositing with adversarial learning. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Long Beach, CA, USA. 3. Tripathi, S., Chandra, S., Agrawal, A., Tyagi, A., Rehg, J.M., and Chari, V. (2019, January 15–20). Learning to generate synthetic data via compositing. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Long Beach, CA, USA. 4. Zhan, F., Huang, J., and Lu, S. (2019). Hierarchy composition gan for high-fidelity image synthesis. arXiv. 5. Zhang, L., Wen, T., Min, J., Wang, J., Han, D., and Shi, J. (2020, January 23–28). Learning object placement by inpainting for compositional data augmentation. Proceedings of the Computer Vision—ECCV 2020: 16th European Conference, Glasgow, UK. Proceedings, Part XIII 16.
|
|