Marco Schouten

I'm a PhD student at the Pioneer Centre for AI (Belongie Lab) in Copenhagen, advised by Dim Papadopoulos and Serge Belongie, working on efficient and controllable generative models. I'm currently a visiting researcher in David Picard's Imagine Lab at École nationale des ponts et chaussées in Paris.

Before my PhD I completed an MSc in Machine Learning at KTH. I've interned at Adobe Research under Tamy Boubekeur on agentic systems. In my free time, I enjoy drawing.

I am on the job market for 2027.

Email Google Scholar GitHub Linkedin

Selected Publications

I have a background in computer science, machine learning, vision, graphics, and robotics. I am specialising in efficient and controllable generative modelling (diffusion models, LLMs, Agents) for 2D and 3D visual data.

HiddenObjects: Diffusion-Distilled Dense Spatial Priors for Object Placement

Marco Schouten, Ioannis Siglidis, Serge Belongie and Dim P. Papadopoulos

WACV, 2027

MeshQuery: Agentic Seam Planning for UV Parametrization

Marco Schouten, Arthur Roullier, Elie Michel, Ruben Wiersma, Axel Paris and Tamy Boubekeur

Preprint, 2026 (Work done at Adobe)

EditJudge-Bench

EditJudge-Bench: Auditing VLM Image-Edit Judges with Synthetic Ground Truth

Jens Parslov, Marco Schouten and Dim P. Papadopoulos

NeurIPS, 2026

Sherlock Your Queries: Learning to Ask the Right Questions for Dialogue-Based Retrieval

Dong Yun*, Marco Schouten* and Dim P. Papadopoulos (* equal contribution)

Preprint, 2025

Instruction-Guided Image Editing

POEM: Precise Object-level Editing via MLLM control

Marco Schouten, M. Onurcan Kaya, Serge Belongie and Dim P. Papadopoulos

SCIA, 2025; CVEU Workshop @ CVPR 2025