3 citations · 5 across the 7 of their papers we have counts for
11 papers
EquiFusion: Kinematics-Agnostic Human Motion Prediction via Equivariant Latent Diffusion
Cecilia Curreli, Florian Hofherr, Dominik Muhle +3
Existing Stochastic 3D Human Motion Prediction models are fundamentally constrained by hard-coding the skeleton kinematics, severely limiting generalization, preventing cross-datas…
VOCA: Visual Odometry with Codec Awareness
Nouri Alexander Hilscher, Mateo de Mayo, Dominik Muhle +2
Camera pose estimation from image streams is a critical component of spatial world models that integrate perception into planning and decision-making. Nearly all Visual Odometry (V…
Dream-to-Recon: Monocular 3D Reconstruction with Diffusion-Depth Distillation from Single Images
Philipp Wulff, Felix Wimbauer, Dominik Muhle +1
Volumetric scene reconstruction from a single image is crucial for a broad range of applications like autonomous driving and robotics. Recent volumetric reconstruction methods achi…
GECO: Geometrically Consistent Embedding with Lightspeed Inference
Regine Hartwig, Dominik Muhle, Riccardo Marin +1
Recent advances in feature learning have shown that self-supervised vision foundation models can capture semantic correspondences but often lack awareness of underlying 3D geometry…
Foundations and Models in Modern Computer Vision: Key Building Blocks in Landmark Architectures
Radu-Andrei Bourceanu, Neil De La Fuente, Jan Grimm +5
This report analyzes the evolution of key design patterns in computer vision by examining six influential papers. The analysis begins with foundational architectures for image reco…
IPFormer: Visual 3D Panoptic Scene Completion with Context-Adaptive Instance Proposals
Markus Gross, Aya Fahmy, Danit Niwattananan +4
Semantic Scene Completion (SSC) has emerged as a pivotal approach for jointly learning scene geometry and semantics, enabling downstream applications such as navigation in mobile r…
Prototype Augmented Hypernetworks for Continual Learning
Neil De La Fuente, Maria Pilligua, Daniel Vidal +4
Continual learning (CL) aims to learn a sequence of tasks without forgetting prior knowledge, but gradient updates for a new task often overwrite the weights learned earlier, causi…
AnyCam: Learning to Recover Camera Poses and Intrinsics from Casual Videos
Felix Wimbauer, Weirong Chen, Dominik Muhle +2
Estimating camera motion and intrinsics from casual videos is a core challenge in computer vision. Traditional bundle-adjustment based methods, such as SfM and SLAM, struggle to pe…
Nonisotropic Gaussian Diffusion for Realistic 3D Human Motion Prediction
Cecilia Curreli, Dominik Muhle, Abhishek Saroha +3
Probabilistic human motion prediction aims to forecast multiple possible future movements from past observations. While current approaches report high diversity and realism, they o…
Boosting Self-Supervision for Single-View Scene Completion via Knowledge Distillation
Keonhee Han, Dominik Muhle, Felix Wimbauer +1
Inferring scene geometry from images via Structure from Motion is a long-standing and fundamental problem in computer vision. While classical approaches and, more recently, depth m…
Cache Me if You Can: Accelerating Diffusion Models through Block Caching
Felix Wimbauer, Bichen Wu, Edgar Schoenfeld +11
Diffusion models have recently revolutionized the field of image synthesis due to their ability to generate photorealistic images. However, one of the major drawbacks of diffusion…