Projects
Open-source
Detectors May 2023
Open source package to accelerate research on out-of-distribution (OOD) detection for computer vision applications.
This library is aimed at assisting researchers in the field of generalized OOD detection. It is inspired by HF's Transformers and features implementations of baselines, metrics, and data sets that allow researchers to perform meaningful benchmarking and development of ood detection methods. It features:
methods: more than 20 detection methods implemented.pipelines: evaluating OOD detectors on popular benchmarks, such as MNIST, CIFAR, and ImageNet benchmarks with random seed support for reproducibility.datasets: OOD datasets implemented with md5 checksums and without the need to download them manually.models: model architectures totally integrated withtimm.eval: implementation of fast OOD evaluation metrics.- Several aggregation methods for multi-layer OOD detection.
- Pipelines for open set recognition and covariate drift detection.
Hackatons
BCG Gamma Deep Learning Hackathon February 2020, São José dos Campos, Brazil
Organized in joint collaboration between the Instituto Tecnológico de Aeronáutica (ITA) and BCG Gamma Brazil. We fine-tuned a YOLOv3 model for high-precision object detection from infrared drone footage receiving the Best Presentation Award. The project was advised by Prof. Filipe Verri.
Aerodynamics
I was involved in research projects in the field of aerospace engineering and aerodynamics theory for airplanes and rocketships being awarded a national scholarship from the CNPq for studying wavy leading edge phenomena with computational fluid dynamics (CFD).
Prompt Riddle
As a fun side project, I co-developed promptriddle, a challenging enigma solving game on an emulated terminal with over 2000 players in our Discord community server.
Volunteering
I volunteered as a high school math teacher at a Brazilian public school, at the social action club at ITA called CASSIS, and at the entrepreneurship league during my bachelors.