Published Apr 7, 2021

NVIDIA’s Shalini De Mello Talks Self-Supervised AI, NeurIPS Successes - Ep. 140

NVIDIA’s Shalini De Mello delves into the transformative potential of self-supervised AI and computer vision, highlighting advancements in automotive safety and 3D reconstruction. Her insights on reducing dependency on large datasets promise a new era of adaptable and fair AI systems.

Episode Highlights

Topics covered

Episode Highlights

Related Episodes

Behind the Scenes at NeurIPS with NVIDIA and CalTech’s Anima Anandkumar - Ep. 131
Answers 383 questions
Ep. 4: How AI Will Revolutionize Driving — Danny Shapiro, NVIDIA
Answers 383 questions
NVIDIA’s Clément Farabet on Orchestrating AI Training for Autonomous Vehicles - Ep. 175
Answers 383 questions
Demystifying AI with NVIDIA’s Will Ramey - Ep. 113
Answers 383 questions
NVIDIA Chief Scientist Bill Dally on Where AI Goes Next - Ep. 62
Answers 383 questions
NVIDIA's Bryan Catanzaro on the Latest from NVIDIA Research - Ep. 57
Answers 383 questions
Ep. 1: Deep Learning 101 - Will Ramey, NVIDIA Senior Manager for GPU Computing
Answers 383 questions
NVIDIA’s Liila Torabi Talks the New Era of Robotics Through Isaac Sim - Ep. 147
Answers 383 questions
NVIDIA Research's David Luebke on Intersection of Graphics, AI - Ep. 127
Answers 383 questions
NVIDIA’s Jim Fan Delves Into Large Language Models and Their Industry Impact - Ep. 204
Answers 383 questions
Ep. 2: Where Deep Learning Goes Next - Bryan Catanzaro, NVIDIA Applied Deep Learning Research
Answers 383 questions
Exploring the AI Startup Ecosystem with NVIDIA Inception’s Jeff Herbst - Ep. 128
Answers 383 questions
NVIDIA’s Annamalai Chockalingam on the Rise of LLMs - Ep. 206
Answers 383 questions
NVIDIA Research's Aaron Lefohn on What's Next at Intersection of AI and Computer Graphics – Ep. 125
Answers 383 questions
NVIDIA’s Sarah Tariq Turning Cars into Software-Defined Vehicles - Ep. 162
Answers 383 questions

Dexa/The AI Podcast

NVIDIA’s Shalini De Mello Talks Self-Supervised AI, NeurIPS Successes - Ep. 140

Topics covered

Popular Clips

Gaze and Image Processing

3D Reconstruction Insights

Passion and Curiosity

Passion for Image Processing

Disentangling Gaze Control

Few Shot Learning

3D Reconstruction Techniques

Learning with Less

Neural Networks Unveiled

Distracted Driving Solutions

Human-Centric Vision Challenges

Mesh Reconstruction Insights

Episode Highlights

Human-Centric Vision

AI Learning Methods

3D Reconstruction

Related Episodes

Behind the Scenes at NeurIPS with NVIDIA and CalTech’s Anima Anandkumar - Ep. 131

Ep. 4: How AI Will Revolutionize Driving — Danny Shapiro, NVIDIA

NVIDIA’s Clément Farabet on Orchestrating AI Training for Autonomous Vehicles - Ep. 175

Demystifying AI with NVIDIA’s Will Ramey - Ep. 113

NVIDIA Chief Scientist Bill Dally on Where AI Goes Next - Ep. 62

NVIDIA's Bryan Catanzaro on the Latest from NVIDIA Research - Ep. 57

Ep. 1: Deep Learning 101 - Will Ramey, NVIDIA Senior Manager for GPU Computing

NVIDIA’s Liila Torabi Talks the New Era of Robotics Through Isaac Sim - Ep. 147

NVIDIA Research's David Luebke on Intersection of Graphics, AI - Ep. 127

NVIDIA’s Jim Fan Delves Into Large Language Models and Their Industry Impact - Ep. 204

Ep. 2: Where Deep Learning Goes Next - Bryan Catanzaro, NVIDIA Applied Deep Learning Research

Exploring the AI Startup Ecosystem with NVIDIA Inception’s Jeff Herbst - Ep. 128

NVIDIA’s Annamalai Chockalingam on the Rise of LLMs - Ep. 206

NVIDIA Research's Aaron Lefohn on What's Next at Intersection of AI and Computer Graphics – Ep. 125

NVIDIA’s Sarah Tariq Turning Cars into Software-Defined Vehicles - Ep. 162

NVIDIA’s Shalini De Mello Talks Self-Supervised AI, NeurIPS Successes - Ep. 140

Topics covered

Popular Clips

Episode Highlights

Human-Centric VisionShalini De Mello, a principal research scientist at NVIDIA, discusses her groundbreaking work in self-supervised AI and computer vision. Her innovations in distracted driving technology and gaze redirection are transforming automotive safety and image processing.

Human-Centric Vision

AI Learning MethodsShalini De Mello explores the intricacies of few-shot and self-supervised learning, highlighting their potential to revolutionize AI by reducing dependency on large datasets. She emphasizes the importance of these techniques in creating more adaptable and fair AI systems.

AI Learning Methods

3D ReconstructionShalini De Mello discusses her groundbreaking work on 3D reconstruction techniques and temporal consistency in computer vision. Her research at NeurIPS focuses on innovative methods for reconstructing 3D shapes from 2D images and maintaining coherence across video frames.

3D Reconstruction

Related Episodes