Complex Lab and Team Project (SS 2025)
Modules
SWS 0/0/4
INF-B-510, INF-B-520, INF-B-530, INF-B-540,
INF-VERT2, INF-VMI-8, INF-VMI-8a, (for INF-VERT7 talk to Prof. Gumhold),
INF-MA-PR, INF-E-4, INF-04-KP, MINF-04-KP-FG1, IST-05-KP
SWS 0/0/8
CMS-VC-TEA (Visual Computing Team Project)
Organisation
UPDATE09.04.25: OPAL course now available for registration.
This year we offer the three topics described below. A minimum of two students are necessary for each topic though.
- We will have our hybrid kickoff meeting on 09.04.2025
- Time - 13:00
- Room - APB 2026
- BBB room https://bbb.tu-dresden.de/rooms/vdd-kvl-uhi-ipq/join
- If you cannot participate in the kickoff meeting email one of the topic responsible people
- After kickoff you can enroll to the topics in the OPAL course.
Topic "Procedural Shader Framework"
Advisor: Benjamin Russig
Procedural shaders allow provide amazing static and dynamic visuals that exploit a wide range of computer graphics techniques like geometric modelling, annimation, and lighting. Some examples can be found on shadertoy: Dolphin, Ladybug, Sea, Clouds, Landscape, ...
Basic Tasks
In this lab the following tasks should be solved in order to make procedural shading accessible inside of graphics applications like game engines:
-
research of sources for procedural shaders beyond shadertoy and collection of a large number of procedural shaders covering as many cg topics as possible
-
analyze shader features and define functional parts (geometry, animation, lighting, after effects, …) with potential programming interfaces
-
re-engineer the collected shaders into functional parts respecting the interfaces and extract parameterizable features with useful parameter ranges
-
Implement re-combination and parameterization framework on different platforms (cgv-framework, shadertoy, unity, unreal, webgl, …)
Requirements
- Solid programming and software engineering skills
- Glsl shader programming skills on the level of CG1
Nice-to-have
- CG knowledge beyond CG1
- Media Design knowledge
Topic "Generating Realistically Clothed Avatars"
Advisor: Kristijan Bartol
Synthetic data plays a significant role in the modeling and understanding of the real world. With the advancements in realistic human body modeling, digital avatars with diverse body shapes have become widespread, especially for metaverse technology. Digital garments, on the other hand, are mostly developed in the gaming industry, where the focus is on the aesthetics and less on the realism of the fabric and fit.
Basic Tasks
In this project, the goal is to
- analysis of the bedlam data set and data generation pipeline (https://bedlam.is.tue.mpg.de/) to place clothed avatars in realistic poses into virtual scenes.
- collect garment models from existing works like Garment Code (https://igl.ethz.ch/projects/GarmentCodeData/)
- drape the garment models onto posed human skin surface models using cloth simulation as for example provided by blender.
- generate new daraset where realistically clothed avatars with varying garment, textures, body shapes, and motion sequences populate high-quality scenes.
Requirements
The main requirements for the successful implementation include:
- experience with any 3D software such as Blender, Unreal, Unity, Houdini, etc.
- basic understanding of articulated bodies and simulations.
Nice-to-have
Ideally, the group should have experience with or interest in virtual avatars or garment modeling:
- rigging and animation of virtual avatars
- experience with cloth simulations and collision management in Blender
- experience with Unreal Engine (general)
Topic "kleintron"
Advisor: Stefan Gumhold
Teaching computer graphics can be supported by interactive material like apps that allow to present graphical algorithms visually and to explore their input and algorithmic parameters interactively. An alternative are computer games that teach cg topics on the fly. In this project the students should develop a Tron like game that can be played on surfaces with different base topologies like sphere, torus or Klein's bottle.
A sample implementation of kleintron implemented in the cgv-framework is provided and can be used as foundation or just for inspirarion.
Basic Tasks
- conception of a game
- implementation of game logic
- game balancing
Requirements:
- CG1 knowledge about the topology of surfaces
- experience with cgv-framework and or other platforms like game engines.
Nice-to-have
- a good spatial cognition and comprehension