Congratulations to the three papers accepted at ECCV 2024!

• FrePolad: Frequency-Rectified Point Latent Diffusion for Point Cloud Generation: a point cloud generation pipeline integrating a variational autoencoder (VAE) with a denoising diffusion probabilistic model (DDPM) for the latent distribution simultaneously achieving high quality, diversity, and flexibility in point cloud cardinality for generation tasks while maintaining high computational efficiency.
• UMBRAE: Unified Multimodal Brain Decoding: performs Unified Cross-Subject Multimodal Brain Decoding using pretrained MLLMs with given prompts providing multimodal explanations to help unveil the ‘Dark Side’ of the brain modality.
• Hypernetworks for Generalizable BRDF Representation: represents material BRDFs as neural continuous functions from their discrete samples using a hypernetwork model with real-world applications that aim for immersive realism offering superior results in both sparse reconstruction and compression.