We propose DiffusionDepth, a new approach that reformulates monocular depth estimation as a denoising diffusion process.

We formulate monocular depth estimation using denoising diffusion models, inspired by their recent successes in high fidelity image generation.

Jul 23, 2024 · We present a novel approach designed to address the complexities posed by challenging, out-of-distribution data in the single-image depth estimation task.

We present a novel approach designed to address the complexities posed by challenging, out-of-distribution data in the single-image depth estimation task.

Mar 9, 2023 · Abstract:Monocular depth estimation is a challenging task that predicts the pixel-wise depth from a single 2D image.

We present Marigold, a diffusion model and associated fine-tuning protocol for monocular depth estimation. Its core principle is to leverage the rich visual ...

We introduce Marigold, a method for affine-invariant monocular depth estimation that is derived from Stable Dif- fusion and retains its rich prior knowledge.

We present a method to estimate dense depth by optimizing a sparse set of points such that their diffusion into a depth map minimizes a multi-view reprojection ...

In the absence of parallax cues, a learning based sin- gle image depth estimation (SIDE) model relies heavily on shading and contextual cues in the image.

We show that they also excel in estimating optical flow and monocular depth, surprisingly without task-specific architectures and loss functions that are ...