Depth Estimation
629 papers with code • 13 benchmarks • 69 datasets
Depth Estimation is the task of measuring the distance of each pixel relative to the camera. Depth is extracted from either monocular (single) or stereo (multiple views of a scene) images. Traditional methods use multi-view geometry to find the relationship between the images. Newer methods can directly estimate depth by minimizing the regression loss, or by learning to generate a novel view from a sequence. The most popular benchmarks are KITTI and NYUv2. Models are typically evaluated according to a RMS metric.
Benchmarks
These leaderboards are used to track progress in Depth Estimation
| Trend | Dataset | Best Model | Paper | Code | Compare |
|---|---|---|---|---|---|
|
|||||
|
|||||
|
|||||
|
|||||
|
|||||
|
|||||
|
|||||
|
|||||
|
|||||
|
|||||
|
|||||
|
|||||
|
Libraries
Use these libraries to find Depth Estimation models and implementations
Cityscapes
KITTI
ScanNet
NYUv2
Matterport3D
Middlebury
SUNCG
TUM RGB-D
Make3D
2D-3D-S
Subtasks
Most implemented papers
High Quality Monocular Depth Estimation via Transfer Learning
Accurate depth estimation from images is a fundamental task in many applications including scene understanding and reconstruction.
Deeper Depth Prediction with Fully Convolutional Residual Networks
iro-cp/FCRN-DepthPrediction
•
•
This paper addresses the problem of estimating the depth map of a scene given a single RGB image.
Unsupervised Monocular Depth Estimation with Left-Right Consistency
mrharicot/monodepth
•
•
CVPR 2017
Learning based methods have shown very promising results for the task of depth estimation in single images.
Vision Transformers for Dense Prediction
isl-org/DPT
•
•
ICCV 2021
We introduce dense vision transformers, an architecture that leverages vision transformers in place of convolutional networks as a backbone for dense prediction tasks.
Towards Robust Monocular Depth Estimation: Mixing Datasets for Zero-shot Cross-dataset Transfer
intel-isl/MiDaS
•
•
In particular, we propose a robust training objective that is invariant to changes in depth range and scale, advocate the use of principled multi-objective learning to combine data from different sources, and highlight the importance of pretraining encoders on auxiliary tasks.
Digging Into Self-Supervised Monocular Depth Estimation
Per-pixel ground-truth depth data is challenging to acquire at scale.
Depth Prediction Without the Sensors: Leveraging Structure for Unsupervised Learning from Monocular Videos
Models and examples built with TensorFlow
Efficient Attention: Attention with Linear Complexities
cmsflash/efficient-attention
•
•
Dot-product attention has wide applications in computer vision and natural language processing.
From Big to Small: Multi-Scale Local Planar Guidance for Monocular Depth Estimation
cogaplex-bts/bts
•
•
We show that the proposed method outperforms the state-of-the-art works with significant margin evaluating on challenging benchmarks.
What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision?
kyle-dorman/bayesian-neural-network-blogpost
•
•
NeurIPS 2017
On the other hand, epistemic uncertainty accounts for uncertainty in the model -- uncertainty which can be explained away given enough data.
