Computer Vision

Fast & Efficient Normalizing Flows and Applications of Image Generative Models

PhD thesis on efficient normalizing flows and applications of image generative models.

Dec 3, 3030

Private IDD: Data Anonymization in the Indian Driving Dataset using Inpainting

Privacy-preserving anonymization of the Indian Driving Dataset using inpainting.

May 7, 7070

R2I-rPPG: A Robust Region of Interest Selection Method for Remote Photoplethysmography to Extract Heart Rate

Robust 3D-landmark ROI selection for remote heart-rate estimation from emergency-department videos.

Oct 21, 21210

ICPR 2024 Competition on Safe Segmentation of Drive Scenes in Unstructured Traffic and Adverse Weather Conditions

ICPR 2024 competition report on safe semantic segmentation in unstructured traffic and adverse weather.

Sep 9, 9090

Inverse of Convolution for Sparse Blind Deconvolution

Inverse of convolution for sparse blind deconvolution.

Jun 1, 1010

Affine-StableSR: Coupling Layer-based Lightweight Autoencoder and Stable Diffusion Model for Image Super-Resolution

Lightweight coupling-layer autoencoder with Stable Diffusion for image super-resolution.

Jan 12, 12120

Adaptation of the Super Resolution SOTA for Art Restoration in Camera Capture Images

Adapting diffusion-based super-resolution models for restoring degraded artwork images.

Sep 21, 21210

C4MTS: Challenge on Categorizing Missing Traffic Signs from Contextual Cues

NCVPRIPG 2023 challenge on detecting missing traffic signs from contextual cues.

Jul 2, 2020

Remote Heart Rate Estimation From Videos for Minority Populations in Emergency Department

Oral abstract on remote heart-rate estimation from emergency-department videos.

Apr 15, 15150

Respiratory Rate Estimation From Videos for Minority Populations in Emergency Department

Oral abstract on remote respiratory-rate estimation from emergency-department videos.

Apr 13, 13130

Automated Seed Quality Testing System using GAN and Active Learning

GAN and batch active learning for automated corn-seed quality testing, with a new 26k-image dataset.

Dec 1, 1010