The department Augmented Vision has 4 accepted papers at the upcoming CVPR 2026 conference taking place from June 3-7, 2026 at the Colorado Convention Center, Denver, USA.
The CVPR conference is the premier international conference in computer vision and pattern recognition.
We are happy to anounce that our work on sensor generalization has won a Best Paper Award at ICPRAM 2026. The conference was held in Marbella, Spain, from March 2nd to 4th.
The Network of Excellence dAIEDGE organized two strategic workshops at the High Performance Embedded Architectures and Compilers Conference HiPEAC 2026, which took place in Krakow, Poland from January 26 to 28. The events brought together leading researchers, industry representatives and European initiatives to discuss the future of distributed, autonomous and trustworthy edge AI systems.
The workshop “The Intelligent Mesh: Edge AI Technology Roadmap for Orchestrating Autonomous Systems with Agentic and Generative AI”, co-organized by Ovidiu Vermesan from SINTEF, Alain Pagani from DFKI, Marcello Coppola from ST Microelectronics and Fabian Chersi from CEA, focused on the next generation of edge AI architectures. Discussions addressed heterogeneous hardware platforms, edge accelerators, neuromorphic approaches and optimized AI frameworks, as well as Small Language Models and Vision Language Models tailored for embedded systems. A central theme was agentic AI at the edge and the vision of an intelligent mesh of autonomous systems capable of collaboration and orchestration. The workshop contributed to shaping a European roadmap for secure, sovereign and scalable edge AI.
The second workshop, “Sustainable and Trustworthy Edge AI for Robotics”, was co-organized by the Networks of Excellence dAIEDGE, euROBIN, ELIAS and ENFIELD. It featured keynote talks by Maximilian Durner from DLR, Jean Marc Bonnefous from TCS, Georgios Spathoulas from NTNU and Eyup Kun from KU Leuven, and included two technical sessions on efficient, regulation-aware and human-centered AI for robotic systems. The workshop concluded with a panel discussion moderated by Alain Pagani on aligning European lighthouse strategies for sustainable, trustworthy and efficient AI.
Together, the two workshops highlighted dAIEDGE’s leading role in fostering collaboration across European AI networks and in advancing a coordinated strategy for next generation edge intelligence.
We are glad to announce that our paper, NURBGen: High-Fidelity Text-to-CAD Generation through LLM-Driven NURBS Modeling, has been accepted at the Fortieth AAAI Conference on Artificial Intelligence (AAAI) 2026 Singapore. Co-led by Muhammad Sadil Khan and Muhammad Usama under the supervision of Didier Stricker and Muhammad Zeshan Afzal, our research introduces the first framework to generate high-fidelity, editable 3D CAD models directly from text by fine-tuning a large language model to produce structured NURBS surface parameters. To overcome the limitations of existing mesh-based or design-history systems, we propose a hybrid symbolic representation that combines untrimmed NURBS with analytic primitives to robustly handle trimmed surfaces and degenerate regions while maintaining token efficiency. Evaluated on our new partABC dataset of 300k annotated CAD components, NURBGen demonstrates strong performance in geometric fidelity and dimensional accuracy, as confirmed by expert evaluations.
On 16 January 2026, Mohammad Minouei successfully defended his PhD entitled ‘Structural Information Extraction from Document Images: Addressing Challenges in Layout Analysis, Table Detection, and Classification’.
The doctoral thesis was carried out in the Augmented Vision department at DFKI under the supervision of Prof. Dr. Didier Stricker.
The examination committee consisted of Prof. Dr. Didier Stricker (RPTU), Prof. Dr. Faisal Shafait (NUST) and Prof. Dr. Leo Van Waveren (RPTU).
