Text-Guided Image Manipulation via Generative Adversarial Network With Referring Image Segmentation-Based Guidance

Article Computer Science, Information Systems

Semantic-aware deidentification generative adversarial networks for identity anonymization

Hyeongbok Kim et al.

Summary: This paper proposes a Semantic-aware De-identification Generative Adversarial Network (SDGAN) model for privacy protection in computer vision. The model considers the semantic information of images and uses an edge-aware graph representation network to extract facial semantic images for effective retention of facial expressions. Compared to other methods, the model significantly improves de-identification performance and generates more reliable and realistic-looking faces.

MULTIMEDIA TOOLS AND APPLICATIONS (2023)

添加到收藏夹

Article Engineering, Electrical & Electronic

Median Stable Clustering and Global Distance Classification for Cross-Domain Person Re-Identification

Zhiqi Pang et al.

Summary: This study proposes an unsupervised cross-domain ReID method based on median stable clustering (MSC) and global distance classification (GDC). By optimizing the model using labeled source domain and unlabeled target domain, the model achieves good generalization in the target domain. Extensive experiments demonstrate the superior performance of MSC-GDC over existing methods.

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY (2022)

添加到收藏夹

Article Telecommunications

Protecting the trust and credibility of data by tracking forgery trace based on GANs

Shuai Xiao et al.

Summary: This study proposes a new method for forgery detection based on GANs, which utilizes gradient information from grayscale images to generate forged traces, and employs focal loss to address imbalanced classification between real and fake samples.

DIGITAL COMMUNICATIONS AND NETWORKS (2022)

添加到收藏夹

Proceedings Paper Acoustics

GENERATIVE ADVERSARIAL NETWORK INCLUDING REFERRING IMAGE SEGMENTATION FOR TEXT-GUIDED IMAGE MANIPULATION

Yuto Watanabe et al.

Summary: This paper proposes a novel generative adversarial network incorporating referring image segmentation for text-guided image manipulation, which effectively distinguishes text-relevant and irrelevant regions to achieve a balance between accurate attribute generation and reconstruction.

2022 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP) (2022)

添加到收藏夹

Proceedings Paper Computer Science, Artificial Intelligence

SEGMENTATION-AWARE TEXT-GUIDED IMAGE MANIPULATION

Tomoki Haruyama et al.

Summary: This paper presents a novel approach to improve text-guided image manipulation performance by introducing foreground-aware and background-aware biases, addressing the issue of modifying undesired parts in images caused by differences in representation ability between text descriptions and images. By integrating an image segmentation network into the generative adversarial network for image manipulation, the effectiveness of the proposed method is demonstrated through comparative experiments with three state-of-the-art methods.

2021 IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING (ICIP) (2021)

添加到收藏夹

Proceedings Paper Computer Science, Artificial Intelligence