Marcos V. Conde's Projects
AI Image SIgnal Processing and Computational Photography - Bokeh Rendering , Reversed ISP Challenge, Model-Based Image Signal Processors via Learnable Dictionaries. Official repo for NTIRE and AIM Challenges
Predicting Molecular Properties. CHAMPS Kaggle competition 2019.
CommonLit Readability Prize 2021. NLP competition hosted at Kaggle: https://www.kaggle.com/c/commonlitreadabilityprize/. Rate the complexity of literary passages for grades 3-12 classroom use
Snakemake ProtoWorkflow for DNA Analysis.
InstructIR: High-Quality Image Restoration Following Human Instructions https://huggingface.co/spaces/marcosv/InstructIR
Conformer and Blind Noisy Students for Improved Image Quality Assessment
Kuzushiji Recognition Kaggle 2019. Build a DL model to transcribe ancient Kuzushiji into contemporary Japanese characters. Opening the door to a thousand years of Japanese culture.
Official repository for Efficient Goal-based Vehicle Motion Prediction baselines. Presented at CVPR and ICRA Workshops 2022, and ITSC conference 2022.
About me
[AAAI 2024] NILUT: Conditional Neural Implicit 3D Lookup Tables for Image Enhancement. Project Website https://mv-lab.github.io/nilut/
RSNA Intracranial Hemorrhage Detection. The challenge is to build an algorithm to detect acute intracranial hemorrhage and its subtypes. The Dataset provided by the Radiological Society of North America (RSNA) and MD.ai.
2020 RSNA-STR Pulmonary Embolism Detection AI Challenge: Classify Pulmonary Embolism cases in chest CT scans. 12th place solution.
Squanchy is a brand new, easy to learn, general purpose, multi-paradigm, compiled programming language. The language is written from scratch (it includes an integrated lexer, parser, code generator etc). I tried to put together Python and Haskell (or at least the main features from both) in Squanchy. In other words you can use it as a scripting language, can do maths (calculus, matrix ...) due to its functional side, work with data ...
This is an official implementation for "Swin Transformer: Hierarchical Vision Transformer using Shifted Windows".
Swin2SR: SwinV2 Transformer for Compressed Image Super-Resolution and Restoration. Advances in Image Manipulation (AIM) workshop ECCV 2022. Try it out! over 3.3M runs https://replicate.com/mv-lab/swin2sr
VQA & VSR Efficiency and Runtime Evaluation -- AIS 2024 Workshop @ CVPR
"Exploring Vision Transformers for Fine-grained Classification" at CVPRW FGVC8
Google Research 3rd YouTube-8M Video Understanding Challenge 2019. Temporal localization of topics within video. International Conference on Computer Vision (ICCV) 2019.