Axel Sauer from the University of Tübingen presented research on scaling Generative Adversarial Networks (GANs) using pretrained representations. The work explores shaping GANs into causal structures, training them up to 40 times faster, and achieving state-of-the-art image synthesis. The presentation mentions "Counterfactual Generative Networks", "Projected GANs", "StyleGAN-XL”, and “StyleGAN-T". Why it matters: Scaling GANs and improving their training efficiency is crucial for advancing image and video synthesis, with implications for various applications in computer vision, graphics, and robotics.
VinAI Research presented research projects focused on advancing image generation and manipulation using GANs and Diffusion Models. The research aims to improve GANs regarding utility, coverage, and output consistency. For Diffusion Models, the work focuses on improving the models’ speed to approach real-time performance and prevent negative social impact of diffusion-based personalized text-to-image generation. Why it matters: This talk indicates ongoing research and development in generative AI in Southeast Asia, an area of growing interest globally.
The paper introduces OmniGen, a unified framework for generating aligned multimodal sensor data for autonomous driving using a shared Bird's Eye View (BEV) space. It uses a novel generalizable multimodal reconstruction method (UAE) to jointly decode LiDAR and multi-view camera data through volume rendering. The framework incorporates a Diffusion Transformer (DiT) with a ControlNet branch to enable controllable multimodal sensor generation, demonstrating good performance and multimodal consistency.
This paper introduces a new Single Domain Generalization (SDG) method called ConDiSR for medical image classification, using channel-wise contrastive disentanglement and reconstruction-based style regularization. The method is evaluated on multicenter histopathology image classification, achieving a 1% improvement in average accuracy compared to state-of-the-art SDG baselines. Code is available at https://github.com/BioMedIA-MBZUAI/ConDiSR.
KAUST computer scientist Mohamed Elhoseiny and his VISION CAIR team developed Creative Walk Adversarial Networks (CWAN) for novel art generation. CWAN learns from existing art styles and deviates using 'random walk deviation' methods. Human evaluators preferred CWAN-generated art compared to other methods like StyleGAN2. Why it matters: The research demonstrates AI's potential as a valuable tool for artists, enabling the creation of unique and meaningful art, and explores more effective emotional language in image captioning.
The paper introduces ScoreAdv, a novel approach for generating natural adversarial examples (UAEs) using diffusion models. It incorporates an adversarial guidance mechanism and saliency maps to shift the sampling distribution and inject visual information. Experiments on ImageNet and CelebA datasets demonstrate state-of-the-art attack success rates, image quality, and robustness against defenses.