Justin C. Johnson

About

Justin C. Johnson is a computer vision researcher whose work bridges the gap between 2D and 3D scene understanding, with a particular focus on enabling machines to perceive and predict human behavior in real-world environments. His most impactful contribution is the development of **Omni3D**, a large-scale benchmark and model for 3D object detection from single images. This work, which has garnered **89 citations** in just one year, addresses a critical bottleneck in the field: the lack of diverse, large-scale datasets for 3D recognition. By providing a unified framework that scales across indoor and outdoor scenes, Omni3D pushes the boundaries of what’s possible for applications in robotics and augmented reality. Johnson also explored the social dimension of AI with **Social GAN**, a generative adversarial network designed to predict socially acceptable human trajectories. This work tackles the inherent multimodality of human motion, a key challenge for autonomous navigation systems. Together, his research demonstrates a commitment to building robust, scalable models that not only see the world in 3D but also understand the nuanced dynamics of human movement.

Research Focus

Key Achievements

Top Papers

1. 1
   Omni3D: A Large Benchmark and Model for 3D Object Detection in the Wild

   89 citations · 2023
2. 2
   Social GAN: Socially Acceptable Trajectories with Generative Adversarial\n Networks

   8 citations · 2018
3. 3
   Omni3D: A Large Benchmark and Model for 3D Object Detection in the Wild

   4 citations · 2022

Key Collaborators

Contact & Links