Picture for Jan-Michael Frahm

Jan-Michael Frahm

MVPSNet: Fast Generalizable Multi-view Photometric Stereo

Add code
May 18, 2023
Figure 1 for MVPSNet: Fast Generalizable Multi-view Photometric Stereo
Figure 2 for MVPSNet: Fast Generalizable Multi-view Photometric Stereo
Figure 3 for MVPSNet: Fast Generalizable Multi-view Photometric Stereo
Figure 4 for MVPSNet: Fast Generalizable Multi-view Photometric Stereo
Viaarxiv icon

LossMix: Simplify and Generalize Mixup for Object Detection and Beyond

Add code
Mar 18, 2023
Viaarxiv icon

A Surface-normal Based Neural Framework for Colonoscopy Reconstruction

Add code
Mar 13, 2023
Figure 1 for A Surface-normal Based Neural Framework for Colonoscopy Reconstruction
Figure 2 for A Surface-normal Based Neural Framework for Colonoscopy Reconstruction
Figure 3 for A Surface-normal Based Neural Framework for Colonoscopy Reconstruction
Figure 4 for A Surface-normal Based Neural Framework for Colonoscopy Reconstruction
Viaarxiv icon

A Practical Stereo Depth System for Smart Glasses

Add code
Nov 19, 2022
Viaarxiv icon

Toward Edge-Efficient Dense Predictions with Synergistic Multi-Task Neural Architecture Search

Add code
Oct 04, 2022
Figure 1 for Toward Edge-Efficient Dense Predictions with Synergistic Multi-Task Neural Architecture Search
Figure 2 for Toward Edge-Efficient Dense Predictions with Synergistic Multi-Task Neural Architecture Search
Figure 3 for Toward Edge-Efficient Dense Predictions with Synergistic Multi-Task Neural Architecture Search
Figure 4 for Toward Edge-Efficient Dense Predictions with Synergistic Multi-Task Neural Architecture Search
Viaarxiv icon

Leveraging Disentangled Representations to Improve Vision-Based Keystroke Inference Attacks Under Low Data

Add code
Apr 05, 2022
Figure 1 for Leveraging Disentangled Representations to Improve Vision-Based Keystroke Inference Attacks Under Low Data
Figure 2 for Leveraging Disentangled Representations to Improve Vision-Based Keystroke Inference Attacks Under Low Data
Figure 3 for Leveraging Disentangled Representations to Improve Vision-Based Keystroke Inference Attacks Under Low Data
Figure 4 for Leveraging Disentangled Representations to Improve Vision-Based Keystroke Inference Attacks Under Low Data
Viaarxiv icon

VPFusion: Joint 3D Volume and Pixel-Aligned Feature Fusion for Single and Multi-view 3D Reconstruction

Add code
Mar 14, 2022
Figure 1 for VPFusion: Joint 3D Volume and Pixel-Aligned Feature Fusion for Single and Multi-view 3D Reconstruction
Figure 2 for VPFusion: Joint 3D Volume and Pixel-Aligned Feature Fusion for Single and Multi-view 3D Reconstruction
Figure 3 for VPFusion: Joint 3D Volume and Pixel-Aligned Feature Fusion for Single and Multi-view 3D Reconstruction
Figure 4 for VPFusion: Joint 3D Volume and Pixel-Aligned Feature Fusion for Single and Multi-view 3D Reconstruction
Viaarxiv icon

ColDE: A Depth Estimation Framework for Colonoscopy Reconstruction

Add code
Nov 19, 2021
Figure 1 for ColDE: A Depth Estimation Framework for Colonoscopy Reconstruction
Figure 2 for ColDE: A Depth Estimation Framework for Colonoscopy Reconstruction
Figure 3 for ColDE: A Depth Estimation Framework for Colonoscopy Reconstruction
Figure 4 for ColDE: A Depth Estimation Framework for Colonoscopy Reconstruction
Viaarxiv icon

Any-Width Networks

Add code
Dec 06, 2020
Figure 1 for Any-Width Networks
Figure 2 for Any-Width Networks
Figure 3 for Any-Width Networks
Figure 4 for Any-Width Networks
Viaarxiv icon

Revisiting the Threat Space for Vision-based Keystroke Inference Attacks

Add code
Sep 12, 2020
Figure 1 for Revisiting the Threat Space for Vision-based Keystroke Inference Attacks
Figure 2 for Revisiting the Threat Space for Vision-based Keystroke Inference Attacks
Figure 3 for Revisiting the Threat Space for Vision-based Keystroke Inference Attacks
Figure 4 for Revisiting the Threat Space for Vision-based Keystroke Inference Attacks
Viaarxiv icon