Photo of Pavlo Melnyk

Pavlo Melnyk

Postdoc

I am a postdoctoral researcher in the division for Computer Vision and Learning Systems (CVL). My research focuses on geometry, equivariant modelling, and applications of AI for Science. Here at CVL, I have earned my PhD in Electrical Engineering with a specialisation in Computer Vision, where I worked on Geometric Deep Learning under the supervision of Michael Felsberg and with funding from WASP.

For more details about my work, please visit pavlomelnyk.com.

Publications

Pavlo Melnyk, Anmar Karmush, Mårten Wadenbäck, Ania Beatriz Rodríguez-Barrera, Johanna Rosen, Michael Felsberg, Jonas Björk, Benchmark Dataset for Catalysis on 2D MXenes (2026)  https://arxiv.org/abs/2606.00794

Pavlo Melnyk, Cuong Le, Urs Waldmann, Per-Erik Forssén, Bastian Wandt, On the Role of Rotation Equivariance in Monocular 2D-to-3D Human Pose Lifting (2026)  https://arxiv.org/abs/2601.13913

Pavlo Melnyk, Michael Felsberg, Kostas Daniilidis,  E(n)- EQUIVARIANT SPHERICAL DECISION SURFACES, GEOMETRY-GROUNDED REPRESENTATION LEARNING AND GENERATIVE MODELING WORKSHOP, GRAM AT ICLR 2026, Proceedings of Machine Learning Research, pp. 416-432, JMLR-JOURNAL MACHINE LEARNING RESEARCH (2026)

Akshay Chandrasekhar, Pavlo Melnyk, Loss-Invariant Projections as Passive Probes of Learned Representations (2026)  https://arxiv.org/abs/2610.06195

Cuong Le, Pavlo Melnyk, Bastian Wandt, Mårten Wadenbäck,  Flow Matching for Probabilistic Monocular 3D Human Pose Estimation, Transactions on Machine Learning Research (2026)

Coworkers CVL

Organisation