Sorry, you need to enable JavaScript to visit this website.

Supplementary Materials for Iterative Self-Improvement of Vision Language Models for Image Scoring and Self-Explanation

The supplementary material contains information on the generated strokes for each sketch,
the filenames of the sketch animation GIFs, the prompts used for the sketches,
and illustrations of the separation process for individual objects.

Additionally, there is a GIFs.zip file inside Supplements.zip.
One can check the generated GIF animations with that.

Supplementary Materials for ICIP 2025

Recent advancements in text-driven 3D content generation highlight several challenges. Surveys show that users often provide simple text inputs while expecting high-quality results. Generating optimal 3D content from minimal prompts is difficult due to the strong dependency of text-to-3D models on input quality. Moreover, the generation process exhibits high variability, often requiring many attempts to meet user expectations, reducing efficiency. To address this, we propose GPT-4V for self-optimization, enhancing generation efficiency and enabling satisfactory results in a single attempt.

Practice for uploading

Relational Representation Distillation Supplementary Material

3D scene understanding is crucial for facilitating seamless interaction between digital devices and the physical world. Real-time capturing and processing of the 3D scene are essential for achieving this seamless integration. While existing approaches typically separate acquisition and processing for each frame, the advent of resolution-scalable 3D sensors offers an opportunity to overcome this paradigm and fully leverage the otherwise wasted acquisition time to initiate processing.

Deep learning in image classification has achieved remarkable success but at the cost of high resource demands. Model compression through automatic joint pruning-quantization addresses this issue, yet most existing techniques overlook a critical aspect: layer correlations. These correlations are essential as they expose redundant computations across layers, and leveraging them facilitates efficient design space exploration. This study employs Graph Neural Networks (GNN) to learn these inter-layer relationships, thereby optimizing the pruning-quantization strategy for the targeted model.

Deep Metric Learning (DML) based on Convolutional Neural Networks (CNNs) is vulnerable to adversarial attacks. Adversarial training, where adversarial samples are generated at each iteration, is one of the prominent defense techniques for robust DML. However, adversarial training increases computational complexity and causes a trade-off between robustness and generalization. This study proposes a lightweight, robust DML framework that learns a non-linear projection to map the embeddings of a CNN into an adversarially robust space.

Deepfake detection is critical in mitigating the societal threats posed by manipulated videos. While various algorithms have been developed for this purpose, challenges arise when detectors operate externally, such as on smartphones, when users take a photo of deepfake images and upload on the Internet. One significant challenge in such scenarios is the presence of Moire patterns, which degrade image quality and confound conventional classification algorithms, including deep neural networks (DNNs). The impact of Moire patterns remains largely unexplored for deepfake detectors.