Papers › Focal-WNet: An Architecture Unifying Convolution and Attention for Depth Estimation

Focal-WNet: An Architecture Unifying Convolution and Attention for Depth Estimation

18 Jul 2022I2CT 2022 7archive 2025-07-28

Gouthamaan Manimaran, Swaminathan J

Extracting depth information from a single RGB image is a fundamental and challenging task in computer vision with wide-ranging applications. This task cannot be solved using traditional methods like multi-view geometry but only by deep learning. Existing methods using convolutional neural nets produce inconsistent and blurry results due to the lack of long-range dependencies. With the recent success of Transformer networks in computer vision, which can process information locally and globally, we leverage this idea to propose a novel architecture named Focal-WNet in this paper. This architecture consists of two separate encoders and a single decoder. The main aim of this network is to learn most monocular depth cues like relative scale, contrast differences, texture gradient, etc. We incorporate focal self-attention instead of vanilla self-attention to reduce the computational complexity of the network. Along with the focal transformer layers, we leverage a convolutional architecture to learn depth cues that cannot be learned by a transformer alone as some cues like occlusion require a local receptive field and are easier for a conv-net to learn. Extensive experiments show that the proposed Focal-WNet achieves competitive results on two challenging datasets.

PaperPDFCode

Code

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

DecoderDepth EstimationMonocular Depth Estimation

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Monocular Depth Estimation KITTI Eigen split Focal-WNet Delta < 1.25 0.926 #47 of 79 Archive leaderboard report
Monocular Depth Estimation KITTI Eigen split Focal-WNet Delta < 1.25^2 0.986 #47 of 79 Archive leaderboard report
Monocular Depth Estimation KITTI Eigen split Focal-WNet Delta < 1.25^3 0.997 #47 of 79 Archive leaderboard report
Monocular Depth Estimation KITTI Eigen split Focal-WNet RMSE 3.076 #47 of 79 Archive leaderboard report
Monocular Depth Estimation KITTI Eigen split Focal-WNet RMSE log 0.120 #47 of 79 Archive leaderboard report
Monocular Depth Estimation KITTI Eigen split Focal-WNet absolute relative error 0.082 #47 of 79 Archive leaderboard report
Monocular Depth Estimation NYU-Depth V2 Focal-WNet Delta < 1.25 0.875 #58 of 85 Archive leaderboard report
Monocular Depth Estimation NYU-Depth V2 Focal-WNet Delta < 1.25^2 0.980 #58 of 85 Archive leaderboard report
Monocular Depth Estimation NYU-Depth V2 Focal-WNet Delta < 1.25^3 0.995 #58 of 85 Archive leaderboard report
Monocular Depth Estimation NYU-Depth V2 Focal-WNet RMSE 0.398 #58 of 85 Archive leaderboard report
Monocular Depth Estimation NYU-Depth V2 Focal-WNet absolute relative error 0.116 #58 of 85 Archive leaderboard report
Monocular Depth Estimation NYU-Depth V2 Focal-WNet log 10 0.048 #58 of 85 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

Absolute Position EncodingsAdamAttentionBPEDense ConnectionsDropoutFocal TransformersLabel SmoothingLayer NormalizationLinear LayerMulti-Head AttentionPosition-Wise Feed-Forward LayerResidual ConnectionSoftmaxTransformer

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections