no code implementations • 22 Mar 2024 • Aalok Patwardhan, Callum Rhodes, Gwangbin Bae, Andrew J. Davison
Given a sequence of images, we can use the per-frame rotation estimates and their uncertainty to perform multi-frame optimisation, achieving robustness and temporal consistency.
1 code implementation • 1 Mar 2024 • Gwangbin Bae, Andrew J. Davison
Despite the growing demand for accurate surface normal estimation models, existing methods use general-purpose dense prediction models, adopting the same inductive biases as other tasks.
no code implementations • 10 Dec 2023 • Kirill Mazur, Gwangbin Bae, Andrew J. Davison
We address this issue with a new image representation which we call a SuperPrimitive.
1 code implementation • 27 Oct 2023 • Oliver Boyne, Gwangbin Bae, James Charles, Roberto Cipolla
Our FOUND approach tackles this, with 4 main contributions: (i) SynFoot, a synthetic dataset of 50, 000 photorealistic foot images, paired with ground truth surface normals and keypoints; (ii) an uncertainty-aware surface normal predictor trained on our synthetic dataset; (iii) an optimization scheme for fitting a generative foot model to a series of images; and (iv) a benchmark dataset of calibrated images and high resolution ground truth geometry.
1 code implementation • 7 Oct 2022 • Gwangbin Bae, Ignas Budvytis, Roberto Cipolla
The depth of each pixel can be propagated to a query pixel, using the predicted surface normal as guidance.
Ranked #35 on Monocular Depth Estimation on NYU-Depth V2
1 code implementation • 5 Oct 2022 • Gwangbin Bae, Martin de La Gorce, Tadas Baltrusaitis, Charlie Hewitt, Dong Chen, Julien Valentin, Roberto Cipolla, Jingjing Shen
Such models are trained on large-scale datasets that contain millions of real human face images collected from the internet.
Ranked #2 on Synthetic Face Recognition on CPLFW (Accuracy metric)
1 code implementation • 3 Oct 2022 • Florian Langer, Gwangbin Bae, Ignas Budvytis, Roberto Cipolla
This combined information is the input to a pose prediction network, SPARC-Net which we train to predict a 9 DoF CAD model pose update.
1 code implementation • CVPR 2022 • Gwangbin Bae, Ignas Budvytis, Roberto Cipolla
To this end, we propose MaGNet, a novel framework for fusing single-view depth probability with multi-view geometry, to improve the accuracy, robustness and efficiency of multi-view depth estimation.
1 code implementation • ICCV 2021 • Gwangbin Bae, Ignas Budvytis, Roberto Cipolla
Experimental results show that the proposed method outperforms the state-of-the-art in ScanNet and NYUv2, and that the estimated uncertainty correlates well with the prediction error.
Ranked #1 on Surface Normal Estimation on NYU-Depth V2