Learning-based sampling for natural image matting

J Tang, Y Aksoy, C Oztireli, M Gross… - Proceedings of the …, 2019 - openaccess.thecvf.com
The goal of natural image matting is the estimation of opacities of a user-defined foreground
object that is essential in creating realistic composite imagery. Natural matting is a …

Transmatting: Enhancing transparent objects matting with transformers

H Cai, F Xue, L Xu, L Guo - European conference on computer vision, 2022 - Springer
Image matting refers to predicting the alpha values of unknown foreground areas from
natural images. Prior methods have focused on propagating alpha values from known to …

Video‐Based Rendering of Dynamic Stationary Environments from Unsynchronized Inputs

T Thonat, Y Aksoy, M Aittala, S Paris… - Computer Graphics …, 2021 - Wiley Online Library
Abstract Image‐Based Rendering allows users to easily capture a scene using a single
camera and then navigate freely with realistic results. However, the resulting renderings are …

Approximate fast foreground colour estimation

M Forte - 2021 IEEE International Conference on Image …, 2021 - ieeexplore.ieee.org
Compositing objects from one image or video onto another background is a classic
computer vision problem. Compositing requires an estimate of the foreground object's …

Light Field Imaging and Reconstruction Tools for Small Objects

C Sathirasethawong - 2024 - unsworks.unsw.edu.au
Inspired by the idea of cataloguing of insects, an automated 3D digitizing system for small
objects is introduced in this thesis. Previous existing systems have complicated equipment …

[PDF][PDF] Supplemental for Video-Based Rendering of Dynamic Stationary Environments from Unsynchronized Inputs

T Thonat, Y Aksoy, M Aittala, S Paris, F Durand… - diglib.eg.org
Discussion. Our adaptation of multi-scale blending to the time domain implicitly assumes
that the observer's perception of temporal transitions is similar to their perception of the …