Daniel Joska
Papers
2
Total Citations
12
H-Index
2
About
Daniel Joska is a researcher at the forefront of computer vision and biomechanics, specializing in the 3D pose estimation of agile animals in natural environments. His work bridges ecology, evolutionary biology, and robotics, with a primary focus on developing non-invasive methods to capture the complex dynamics of high-speed wildlife. Joska’s major contributions include the creation of **AcinoSet**, a pioneering 3D pose estimation dataset and baseline model for cheetahs in the wild (8 citations), which provides critical data for understanding extreme animal agility and inspiring next-generation legged robots. He further advanced the field by addressing key challenges in **improving 3D markerless pose estimation using low-cost cameras** (4 citations), tackling issues like unnatural pose estimates to enable robust tracking of fast-moving animals. By combining affordable hardware with sophisticated algorithms, Joska’s work makes high-fidelity motion capture accessible for field studies, directly impacting both biological research and robotic controller design. His innovative approach to solving real-world tracking problems positions him as a key contributor to the growing intersection of animal behavior analysis and autonomous systems.
Research Focus
Key Achievements
Top Papers
- 1
- 2