genai4good

GenAI4Good Open Initiative

The tremendous recent advances in generative artificial intelligence techniques have led to significant successes and promise in a wide range of different applications ranging from conversational agents and textual content generation to voice and visual synthesis. Amid the rise in generative AI and its increasing widespread adoption, there has been significant growing concern over the use of generative AI for malicious purposes ranging from disinformation creation to content forgery to cyberattack generation. However, despite the potential for negative socioeconomic impact, generative AI holds tremendous promise in improving the human condition in a large number of ways if created and deployed in a reponsible and trustworthy manner.

Motivated to tackle such challenges and pave the way for positive impact, we have launched GenAI4Good, a global open source initiative for accelerating machine learning for promoting responsible creation and deployment of generative AI for good. The sub-initiatives that are part of GenAI4Good will be listed and updated in this portal over time as the effort grows. GenAI4Good is a joint initiative with the COVID-Net initiative and the Cancer-Net initiative.

AI Illuminated: Generative AI Podcast on the Latest Research Developments

Podcast: Click here
NAS-NeRF

NAS-NeRF: Generative Neural Architecture Search for Neural Radiance Field

Repo: Click here
Paper: Click here

NAS-NeRF

DeepfakeArt Challenge Benchmark Dataset for Generative AI Art Forgery and Data Poisoning Detection

Repo: Click here
Benchmark dataset: Click here

DeepfakeArt

MetaGraspNetV2: All-in-One Dataset Enabling Fast and Reliable Robotic Bin Picking via Object Relationship Reasoning and Dexterous Grasping

Paper: Click here
Benchmark dataset: Click here

meta1 meta1

MetaGraspNet: A Large-Scale Benchmark Dataset for Scene-Aware Ambidextrous Bin Picking via Physics-based Metaverse Synthesis

Paper: Click here
Benchmark dataset: Click here

Nutritionverse3d

NutritionVerse: Large-scale Dataset of Real and Synthetic Meal Data for Nutritional Intake Estimation

Paper: Click here
Benchmark dataset: Click here

Nutritionverse3d

NutritionVerse-3D: A 3D Food Model Dataset for Nutritional Intake Estimation

Paper: Click here
Benchmark dataset: Click here

Nutritionverse3d

FoodFusion: A Latent Diffusion Model for Realistic Food Image Generation

Paper: Click here
Foodfusion

ProstateGAN: Mitigating Data Bias via Prostate Diffusion Imaging Synthesis with Generative Adversarial Networks

Paper: Click here
Nutritionverse3d

Core GenAI4Good Team

Project Lead: Alexander Wong (a28wong@uwaterloo.ca)

Partners: