Roboflow

Depth Estimation Models

The one depth estimation model in our catalog.

1 model · 1 open-weight · prices synced Sep 12, 2026

Open-Source Depth Estimation Models

1 model with downloadable weights you can self-host under their licenses (Apache 2.0).

ByteDance
Depth Anything V2
Depth Anything V2 is a monocular depth estimation model released in June 2024 by researchers at the University of Hong Kong and TikTok. It predicts a dense depth map from a single RGB image, enabling 3D-aware applications without the need for stereo cameras, LiDAR, or multi-view inputs. The model improves on the original Depth Anything through three modifications: replacing real labeled images with 595K high-quality synthetic images during teacher training, scaling up teacher model capacity, and using the stronger teacher to generate pseudo-labels on 62 million unlabeled real images used to train the student models. This pipeline reduces the depth prediction artifacts that can occur in reflective, transparent, and texture-poor regions. Compared to diffusion-based depth models such as Marigold, Depth Anything V2 runs more than 10× faster while producing more accurate predictions.Depth Anything V2 is released in four sizes: Small (25M), Base (97M), Large (335M), and Giant (1.3B), and in two output modes: relative depth (normalized scene-level estimates) and metric depth (absolute distance in meters, produced by fine-tuning the relative-depth backbone on depth-annotated datasets). The Small, Base, and Large model weights are released under Apache 2.0, and the Giant variant under CC-BY-NC-4.0 for non-commercial use. A successor model, Depth Anything 3, was released in November 2025 by the ByteDance Seed team, extending the framework to multi-view depth estimation and camera pose recovery.

How to Choose a Depth Estimation Model

Depth models predict distance per pixel, and the deciding question is whether you need relative depth (what is nearer) or metric depth (how far, in units), because those are different problems.

Monocular models in the MiDaS and Depth Anything lineage generalize across scene types and produce excellent relative depth maps: correct ordering and clean boundaries. That is enough for background blur, AR occlusion, image editing, and scene understanding, and it works with any ordinary camera.

Real distances require metric depth models (trained for absolute scale, often per-domain), stereo cameras, or depth sensors. Monocular metric estimates carry real error that varies by scene, so navigation, measurement, and safety systems should verify against ground truth in their own environment before trusting any model output.

The bottom line: For ordering and visual effects, a monocular relative-depth model is enough. For real distances, use a metric model validated in your environment, or add stereo or a sensor.

Frequently Asked Questions About Depth Estimation Models

Depth estimation is the task of predicting how far each pixel in an image is from the camera, turning a flat photo into a 2.5D understanding of the scene. Monocular models infer depth from a single image using learned priors and may output relative depth (ordering) or metric depth (real distances); stereo and multi-view methods compute it geometrically from parallax. Models in the MiDaS and Depth Anything lineage train across many datasets to generalize broadly, and accuracy is measured by error against sensor ground truth. Depth enables robot and drone navigation, AR occlusion that looks correct, portrait background blur, and scene layout analysis without special hardware. This page covers the one depth estimation model we track, Depth Anything V2, open-weight and free to self-host.

Depth Anything V2 is the only depth estimation model in our catalog, so there is nothing here to rank it against. For fixed categories in production, a model fine-tuned on your own data typically beats any general-purpose model. Try it on your own images to see whether it fits.

Yes. The one depth estimation model we track, Depth Anything V2, is open-weight, free to self-host under its license (Apache 2.0).

This page lists the one depth estimation models in the Roboflow Playground catalog, all of them open-weight and free to self-host. Compare licenses, parameters, and release dates side by side, or open any model page for full details.