Hi, I'm Victor Hoffmann, Computer Vision Research Intern at Sony and AI Student at Institut Polytechnique de Paris.

🔭 Research Projects in Laboratories:

Lorraine Research Laboratory in Computer Science and its Applications (LORIA)
- Deep Learning Approaches for Dynamic Mechanical Analysis of Viscoelastic Fiber Composites
National Institute for Research in Digital Science and Technology (INRIA)
- Machine Learning Techniques for Fragmentation Problems

✍️ Personal Projects:

Real-Time Room Depth Estimation With Monocular Camera
- MiDaS for Real-Time Depth Estimation
Vehicle Recognition for Smart Cities
- YoloV8 and ByteTrack for Traffic Management
Real-Time Segmentation for Biomedical Imaging
- FastSAM for Real-Time MRI Brain Scans (Epilepsy Warning)

🌎 Contribution to Open-Source Communities:

Implementation of a Streaming Graph Anomaly Detection Algorithm on River, a Python Online Learning API
- Implementation of AnoGraph for River API

📄 Article Implementations:

Rainbow: Combining Improvements in Deep Reinforcement Learning Implementation
- Rainbow Agent on Atari Games
CausalVAE: Structured Causal Disentanglement in Variational Autoencoder Implementation
- CausalVAE on Flow and Pendulum

👨‍💻 Data Challenges & Hackathons:

Hi!ckathon 4th edition (2023), Hi! Paris (Won Capgemini’s Best Scientific Approach Award)
- Team 5, Hi!ckathon 4th edition, Hi! Paris
Real Estate Price Prediction (2022), Louis Bachelier Institute, ENS Data Challenge
- Real Estate Price Prediction, Data Challenge
Return Forecasting of Cryptocurrency Clusters (2022), Napoleon X, ENS Data Challenge
- Return Forecasting of Cryptocurrency Clusters, Data Challenge

🧰 Languages and Tools:

🤳 Connect with me on LinkedIn:

@Victor-Hoffmann-AI

Victor Hoffmann's Projects

2023-assignment-numpy

Assignment 1 for the DataCamp course X-DataScience Master -- numpy

2023-assignment-pandas

Assignment 2 for the DataCamp course X-DataScience Master -- pandas

2023-assignment-sklearn

Assignment 3 for the DataCamp course X-DataScience Master - scikit-learn API

datacamp-master

Datacamp class for master student - 1 week

ens-challenge-data-cryptocurrency-clusters

This challenge organized by ENS Ulm and Collège de France was about predicting mean return of cluster's assets relatively to the bitcoin during the last hour of the day, given the last 23 hours.

ens-challenge-data-real-estate

This challenge organized by ENS Ulm and Collège de France was about predicting real estate prices with tabular data and images.

molecule-retrieval-with-natural-language-queries

The goal of this project is to study and apply machine learning/artificial intelligence techniques to retrieve molecules (graphs) using natural language queries. Natural language and molecules encode information in very different ways, which leads to the exciting but challenging problem of integrating these two very different modalities.