Pixelpiece3

Implementation of a Diffusion Transformer (DiT) specifically tuned for depth map synthesis.

This paper explores the transition from latent-space diffusion models to pixel-space diffusion generation . We address the "flying pixel" artifact—a common byproduct of Variational Autoencoder (VAE) compression—by performing diffusion directly in the pixel domain. By leveraging semantics-prompted diffusion , our approach ensures high-quality point cloud reconstruction from single-view images. 1. Introduction Pixelpiece3

Traditional monocular depth models like Marigold often suffer from blurry edges and depth artifacts due to the lossy nature of VAEs. Pixelpiece3

Detailed analysis of how bypassing latent-space compression removes "flying pixels" at depth discontinuities. 3. Quantitative and Qualitative Evaluation Pixelpiece3