Dr. Dimitri(o)s Mallis

Computer Vision Researcher @ SnT • PhD from the University of Nottingham.

Welcome!! 👋 I am a Postdoctoral Researcher at the Interdisciplinary Centre for Security, Reliability and Trust (SnT) of the University of Luxembourg, focusing on computer vision its appliations for Computer-Aided Design (CAD).

I received my PhD in Computer Vision at the University of Nottingham under the supervision of Dr. Yorgos Tzimiropoulos. I was also a Research Intern at Samsung AI Cambridge (SAIC) and later a Senior Machine Learning Engineer for Taboola and Deeplab.

In this website you can find more info about me, check publications and recent news. You can also browse through my photography here 📸 !!

Research

During my PhD, I concentrated on learning with minimal manual supervision through self-supervised and semi-supervised learning, specifically for the tasks of landmark detection and human pose estimation. Later, I also looked into deep learning based recommender systems and the problem of click-through-rate (CTR) prediction.

Currently my focus in on 3D vision and it’s application towards accelerating parametric Computer-Aided Design (CAD). I am particularly interested in automated reverse engineering of 2D/3D CAD models as well as Multimodal Large Language Models (MLLMs) for AI-assisted design. I also lead Morfis, an SnT project focused on AI-assisted physical product customization.

News

Feb 03, 2025	Very exciting news, the Morfis project has been selected to be part of this years cohort of the Venture Program of SnT. Check out the project video here 🎥!
Dec 03, 2024	Paper announcement 📢📢 !!! Our work entitled DAVINCI: A Single-Stage Architecture for Constrained CAD Sketch Inference was presented as an oral 🎉🎉 at the British Machine Vision Conference (BMVC) 2024! For more details check out the project page.
Dec 01, 2024	Our recent work PICASSO: A Feed-Forward Framework for Parametric Inference of CAD Sketches via Rendering Self-Supervision has been accepted for publication at WACV 2025!

Selected Publications [View All]

CAD-Assistant: Tool-Augmented VLLMs as Generic CAD Task Solvers

Dimitrios Mallis, Ahmet Serdar Karadeniz, Sebastian Cavada, Danila Rukhovich, Niki Foteinopoulou, Kseniya Cherenkova, Anis Kacem, and Djamila Aouada

ArXiv, 2025

Abs arXiv Bib

We propose CAD-Assistant, a general-purpose CAD agent for AI-assisted design. Our approach is based on a powerful Vision and Large Language Model (VLLM) as a planner and a tool-augmentation paradigm using CAD-specific tools. CAD-Assistant addresses multimodal user queries by generating actions that are iteratively executed on a Python interpreter equipped with the FreeCAD software, accessed via its Python API. Our framework is able to assess the impact of generated CAD commands on geometry and adapts subsequent actions based on the evolving state of the CAD design. We consider a wide range of CAD-specific tools including a sketch image parameterizer, rendering modules, a 2D cross-section generator, and other specialized routines. CAD-Assistant is evaluated on multiple CAD benchmarks, where it outperforms VLLM baselines and supervised task-specific methods. Beyond existing benchmarks, we qualitatively demonstrate the potential of tool-augmented VLLMs as general-purpose CAD solvers across diverse workflows.
@article{Mallis2025CADAssistant, title = {CAD-Assistant: Tool-Augmented VLLMs as Generic CAD Task Solvers}, author = {Mallis, Dimitrios and Karadeniz, Ahmet Serdar and Cavada, Sebastian and Rukhovich, Danila and Foteinopoulou, Niki and Cherenkova, Kseniya and Kacem, Anis and Aouada, Djamila}, journal = {ArXiv}, year = {2025}, }
CAD-Recode: Reverse Engineering CAD Code from Point Clouds

Danila Rukhovich, Elona Dupont, Dimitrios Mallis, Kseniya Cherenkova, Anis Kacem, and Djamila Aouada

ArXiv, 2025

Abs arXiv Bib URL Code

Computer-Aided Design (CAD) models are typically constructed by sequentially drawing parametric sketches and applying CAD operations to obtain a 3D model. The problem of 3D CAD reverse engineering consists of reconstructing the sketch and CAD operation sequences from 3D representations such as point clouds. In this paper, we address this challenge through novel contributions across three levels: CAD sequence representation, network design, and training dataset. In particular, we represent CAD sketch-extrude sequences as Python code. The proposed CAD-Recode translates a point cloud into Python code that, when executed, reconstructs the CAD model. Taking advantage of the exposure of pre-trained Large Language Models (LLMs) to Python code, we leverage a relatively small LLM as a decoder for CAD-Recode and combine it with a lightweight point cloud projector. CAD-Recode is trained on a procedurally generated dataset of one million CAD sequences. CAD-Recode significantly outperforms existing methods across the DeepCAD, Fusion360 and realworld CC3D datasets. Furthermore, we show that our CAD Python code output is interpretable by off-the-shelf LLMs, enabling CAD editing and CAD-specific question answering from point clouds.
author = {Rukhovich, Danila and Dupont, Elona and Mallis, Dimitrios and Cherenkova, Kseniya and Kacem, Anis and Aouada, Djamila}, journal = {ArXiv}, year = {2025}, }
DAVINCI: A Single-Stage Architecture for Constrained CAD Sketch Inference

Ahmet Serdar Karadeniz, Dimitrios Mallis, Nesryne Mejri, Kseniya Cherenkova, Anis Kacem, and Djamila Aouada

British Machine Vision Conference (BMVC), 2024

Abs arXiv Bib URL Code

This work presents DAVINCI, a unified architecture for single-stage Computer-Aided Design (CAD) sketch parameterization and constraint inference directly from raster sketch images. By jointly learning both outputs, DAVINCI minimizes error accumulation and enhances the performance of constrained CAD sketch inference. Notably, DAVINCI achieves state-of-the-art results on the large-scale SketchGraphs dataset, demonstrating effectiveness on both precise and hand-drawn raster CAD sketches. To reduce DAVINCI’s reliance on large-scale annotated datasets, we explore the efficacy of CAD sketch augmentations. We introduce Constraint-Preserving Transformations (CPTs), i.e. random permutations of the parametric primitives of a CAD sketch that preserve its constraints. This data augmentation strategy allows DAVINCI to achieve reasonable performance when trained with only 0.1% of the SketchGraphs dataset. Furthermore, this work contributes a new version of SketchGraphs, augmented with CPTs. The newly introduced CPTSketchGraphs dataset includes 80 million CPT-augmented sketches, thus providing a rich resource for future research in the CAD sketch domain.
@article{Karadeniz2024PICASSOAF, title = {DAVINCI: A Single-Stage Architecture for Constrained CAD Sketch Inference}, author = {Karadeniz, Ahmet Serdar and Mallis, Dimitrios and Mejri, Nesryne and Cherenkova, Kseniya and Kacem, Anis and Aouada, Djamila}, journal = {British Machine Vision Conference (BMVC)}, year = {2024}, }
From Keypoints to Object Landmarks via Self-Training Correspondence: A novel approach to Unsupervised Landmark Discovery

Dimitrios Mallis, Enrique Sanchez, Matt Bell, and Georgios Tzimiropoulos

IEEE Transactions on Pattern Analysis and Machine Intelligence, 2023

Abs arXiv Bib URL Code

This paper proposes a novel paradigm for the unsupervised learning of object landmark detectors. Contrary to existing methods that build on auxiliary tasks such as image generation or equivariance, we propose a self-training approach where, departing from generic keypoints, a landmark detector and descriptor is trained to improve itself, tuning the keypoints into distinctive landmarks. To this end, we propose an iterative algorithm that alternates between producing new pseudo-labels through feature clustering and learning distinctive features for each pseudo-class through contrastive learning. With a shared backbone for the landmark detector and descriptor, the keypoint locations progressively converge to stable landmarks, filtering those less stable. Compared to previous works, our approach can learn points that are more flexible in terms of capturing large viewpoint changes. We validate our method on a variety of difficult datasets, including LS3D, BBCPose, Human3.6M and PennAction, achieving new state of the art results.
@article{Mallis2023landm, author = {Mallis, Dimitrios and Sanchez, Enrique and Bell, Matt and Tzimiropoulos, Georgios}, journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence}, title = {From Keypoints to Object Landmarks via Self-Training Correspondence: A novel approach to Unsupervised Landmark Discovery}, year = {2023}, }
Unsupervised Learning of Object Landmarks via Self-Training Correspondence

Dimitrios Mallis, Enrique Sanchez, Matt Bell, and Georgios Tzimiropoulos

Neural Information Processing Systems, 2020

Abs Bib URL PDF Supp Code

This paper addresses the problem of unsupervised discovery of object landmarks. We take a different path compared to existing works, based on 2 novel perspectives: (1) Self-training: starting from generic keypoints, we propose a self-training approach where the goal is to learn a detector that improves itself, becoming more and more tuned to object landmarks. (2) Correspondence: we identify correspondence as a key objective for unsupervised landmark discovery and propose an optimization scheme which alternates between recovering object landmark correspondence across different images via clustering and learning an object landmark descriptor without labels. Compared to previous works, our approach can learn landmarks that are more flexible in terms of capturing large changes in viewpoint. We show the favourable properties of our method on a variety of difficult datasets including LS3D, BBCPose and Human3.6M.
@article{unsupervLandm2020, title = {Unsupervised Learning of Object Landmarks via Self-Training Correspondence}, author = {Mallis, Dimitrios and Sanchez, Enrique and Bell, Matt and Tzimiropoulos, Georgios}, journal = {Neural Information Processing Systems}, year = {2020}, }

Research

News

Selected Publications [View All]

Open Source Projects