6 papers
Prompting Image Generators for Training-free Primitive Shape Abstraction
Gregor Kobsik, Tim Elsner, Leif Kobbelt
Compact primitive abstractions represent 3D shapes with a few geometric primitives while preserving recognizable components. Learned methods depend on their training classes, and o…
Learning Fine-to-Coarse Cuboid Shape Abstraction
Gregor Kobsik, Morten Henkel, Yanjiang He +4
The abstraction of 3D objects with simple geometric primitives like cuboids allows to infer structural information from complex geometry. It is important for 3D shape understanding…
Multidimensional Byte Pair Encoding: Shortened Sequences for Improved Visual Data Generation
Tim Elsner, Paula Usinger, Julius Nehring-Wirxel +5
In language processing, transformers benefit greatly from text being condensed. This is achieved through a larger vocabulary that captures word fragments instead of plain character…
Quantised Global Autoencoder: A Holistic Approach to Representing Visual Data
Tim Elsner, Paula Usinger, Victor Czech +4
In quantised autoencoders, images are usually split into local patches, each encoded by one token. This representation is redundant in the sense that the same number of tokens is s…
Partial Symmetry Detection for 3D Geometry using Contrastive Learning with Geodesic Point Cloud Patches
Gregor Kobsik, Isaak Lim, Leif Kobbelt
Detecting partial extrinsic symmetry in 3D geometry is a fundamental yet persistent challenge in computer vision and graphics, critical for tasks ranging from shape completion to p…
Octree Transformer: Autoregressive 3D Shape Generation on Hierarchically Structured Sequences
Moritz Ibing, Gregor Kobsik, Leif Kobbelt
Autoregressive models have proven to be very powerful in NLP text generation tasks and lately have gained popularity for image generation as well. However, they have seen limited u…