papers

Publications (8)

cs.CV2023

MotionDeltaCNN: Sparse CNN Inference of Frame Differences in Moving Camera Videos

Mathias Parger, Chengcheng Tang, Thomas Neff +4

Convolutional neural network inference on video input is computationally expensive and requires high memory bandwidth. Recently, DeltaCNN managed to reduce the cost by only process…

cs.CV2020

UNOC: Understanding Occlusion for Embodied Presence in Virtual Reality

Mathias Parger, Chengcheng Tang, Yuanlu Xu +5

Tracking body and hand motions in the 3D space is essential for social and self-presence in augmented and virtual environments. Unlike the popular 3D pose estimation setting, the p…

cs.CV2024

Collaborative Control for Geometry-Conditioned PBR Image Generation

Shimon Vainer, Mark Boss, Mathias Parger +5

Graphics pipelines require physically-based rendering (PBR) materials, yet current 3D content generation approaches are built on RGB models. We propose to model the PBR image distr…

cs.LG2022

Gradient-based Weight Density Balancing for Robust Dynamic Sparse Training

Mathias Parger, Alexander Ertl, Paul Eibensteiner +3

Training a sparse neural network from scratch requires optimizing connections at the same time as the weights themselves. Typically, the weights are redistributed after a predefine…

cs.LG2026

Terminal Velocity Matching

Linqi Zhou, Mathias Parger, Ayaan Haque +1

We propose Terminal Velocity Matching (TVM), a generalization of flow matching that enables high-fidelity one- and few-step generative modeling. TVM models the transition between a…

cs.CV2021

DONeRF: Towards Real-Time Rendering of Compact Neural Radiance Fields using Depth Oracle Networks

Thomas Neff, Pascal Stadlbauer, Mathias Parger +5

The recent research explosion around implicit neural representations, such as NeRF, shows that there is immense potential for implicitly storing high-quality scene and lighting inf…