Project page of the paper "Learning Multi-Scale Photo Exposure Correction" (CVPR 2021).
- deep-learning
- deeplearning
- multi-scale
- coarse-to-fine
142,273 visible repositories in the index
73–96 of 97 · Most stars
Project page of the paper "Learning Multi-Scale Photo Exposure Correction" (CVPR 2021).
A physics-based lighting system for CSS. Define a light source, and every shadow, highlight and surface gradient follows from it — calibrated against Blender raytraces.
Scholar All-In-One: A research infrastructure for AI agents
Official repository for the paper "Grokfast: Accelerated Grokking by Amplifying Slow Gradients"
Physics-based Noise Modeling for Extreme Low-light Photography (CVPR 2020 Oral & TPAMI 2021)
A TensorFlow implementation of DeepMind's Differential Neural Computers (DNC)
U-Net Brain Tumor Segmentation
Exascale multiphase flow solver — 2025 Gordon Bell Prize Finalist | 200T grid points on 43K+ GPUs
Neural networks toolbox focused on medical image analysis
A lightweight library to calculate tensors in Swift, which has similar APIs to TensorFlow's
City Energy Analyst (CEA) is an open-source urban building energy modeling (UBEM) platform and computation tool for the design of low-carbon and highly efficient cities.
Kaapana is an open source toolkit for state of the art platform provisioning in the field of medical data analysis. The applications comprise AI-based workflows and federated learning scenarios with a focus on radiological and radiotherapeutic imaging. The name Kaapana comes from the Hawaiian word kaʻāpana, meaning "distributor" or "part".
Bayesian Optimization using GPflow
An ESP32 and 4.2" ePaper Display reads Dark Sky weather API and displays the weather using ESP-IDF
Accurate and efficient spot detection for microscopy data
⚡ Open-source software for deep learning-based digital pathology
Auto-QChem is an automated workflow for the generation and storage of DFT calculations for organic molecules.
Learning cell communication from spatial graphs of cells
C# math expression library with symbolic computation support
2D triangulation library. Allows translating lines and polygons (both based on points) to the language of GPUs.
Multi-stage Riemannian flow matching for physically valid molecular docking, with GNINA scoring, PoseBusters filtering, CLI inference, and benchmarks.
UV-first benchmark for protein-ligand docking with reproducible annotation, pocket similarity, and HPC workflows.
This is a class module in beta state to implement computations with large integers and reals/rationals numbers, while proposing representation of numerical values as scientific notation, allowing to obtain several cohorts for the same number (a lot of fun!).
Functions for the input/output and visualization of medical imaging data that follow either the ANALYZE, NIfTI or AFNI formats.