Marco Schouten

I am a PhD candidate at DTU and the Pioneer Centre for AI, advised by Prof. Dim P. Papadopoulos and Prof. Serge Belongie (Belongie Lab). My research focuses on generative modeling: controllable diffusion for image editing, structured scene understanding, and multimodal/VLM-guided editing.

Previously, research intern at Aptiv (HD map generation, autonomous driving) and the Swedish Maritime Robotics Centre (SLAM, underwater robotics).

Outside work I enjoy drawing

Open to research internship opportunities for 2026.

Publications

HiddenObjects: Diffusion-Distilled Dense Spatial Priors for Object Placement Marco Schouten, Ioannis Siglidis, Serge Belongie, Dim P. Papadopoulos Under Review, 2026

Instruction-Guided Precise Image Editing Using Multimodal LLMs Marco Schouten, M. Onurcan Kaya, Serge Belongie, Dim P. Papadopoulos CVEU @ CVPR Workshop, 2025 — PDF

POEM: Precise Object-Level Editing via MLLM Control Marco Schouten, M. Onurcan Kaya, Serge Belongie, Dim P. Papadopoulos Scandinavian Conference on Image Analysis (SCIA), 2025 — webpage

Sherlock Your Queries: Learning to Ask the Right Questions for Dialogue-Based Retrieval Dong Yun*, Marco Schouten*, Dim P. Papadopoulos (* equal contribution) arXiv, 2025 — arXiv

Service & Honors

Reviewer, IJCV (2026) · Selected, ICVSS (2025) · Hattrick Award, KTH (2020) · Selected, LeadTheFuture (2020) · Scholarship, University of Genoa (2018–2020)