Thomas Verelst

11 papers A* 2A 1C 1Journal 6Unranked 1
YearRankTypeTitle / Venue / Authors
2023 J jnl
SegBlocks: Block-Based Dynamic Resolution Networks for Real-Time Segmentation.
IEEE Trans. Pattern Anal. Mach. Intell.
Thomas Verelst, Tinne Tuytelaars
2023 A conf
Spatial Consistency Loss for Training Multi-Label Classifiers from Single-Label Annotations.
WACV
Thomas Verelst, Paul K. Rubenstein, Marcin Eichner, Tinne Tuytelaars, Maxim Berman
2022 J jnl
Spatial Consistency Loss for Training Multi-Label Classifiers from Single-Label Annotations.
CoRR
Thomas Verelst, Paul K. Rubenstein, Marcin Eichner, Tinne Tuytelaars, Maxim Berman
2021 A* conf
BlockCopy: High-Resolution Video Processing with Block-Sparse Feature Propagation and Online Policies.
ICCV
Thomas Verelst, Tinne Tuytelaars
2021 J jnl
BlockCopy: High-Resolution Video Processing with Block-Sparse Feature Propagation and Online Policies.
CoRR
Thomas Verelst, Tinne Tuytelaars
2021 C conf
Processor Architecture Optimization for Spatially Dynamic Neural Networks.
VLSI-SoC
Steven Colleman, Thomas Verelst, Linyan Mei, Tinne Tuytelaars, Marian Verhelst
2020 A* conf
Dynamic Convolutions: Exploiting Spatial Sparsity for Faster Inference.
CVPR
Thomas Verelst, Tinne Tuytelaars
2020 J jnl
SegBlocks: Block-Based Dynamic Resolution Networks for Real-Time Segmentation.
CoRR
Thomas Verelst, Tinne Tuytelaars
2020 — conf
SegBlocks: Towards Block-Based Adaptive Resolution Networks for Fast Segmentation.
ECCV Workshops (5)
Thomas Verelst, Tinne Tuytelaars
2019 J jnl
Dynamic Convolutions: Exploiting Spatial Sparsity for Faster Inference.
CoRR
Thomas Verelst, Tinne Tuytelaars
2019 J jnl
Generating superpixels using deep image representations.
CoRR
Thomas Verelst, Matthew B. Blaschko, Maxim Berman