3 papers
cs.CV2026
Efficient Image Synthesis with Sphere Latent Encoder
Tung Do, Thuan Hoang Nguyen, Hao Li
Few-step image generation has seen rapid progress, with consistency and meanflow-based methods significantly reducing the number of sampling steps. Despite their low inference cost…
cs.CV2024
LiftRefine: Progressively Refined View Synthesis from 3D Lifting with Volume-Triplane Representations
Tung Do, Thuan Hoang Nguyen, Anh Tuan Tran +2
We propose a new view synthesis method via synthesizing a 3D neural field from both single or few-view input images. To address the ill-posed nature of the image-to-3D generation p…
cs.CV2024
SharpDepth: Sharpening Metric Depth Predictions Using Diffusion Distillation
Duc-Hai Pham, Tung Do, Phong Nguyen +3
We propose SharpDepth, a novel approach to monocular metric depth estimation that combines the metric accuracy of discriminative depth estimation methods (e.g., Metric3D, UniDepth)…