SimpleNeRF: Regularizing Sparse Input Neural Radiance Fields with Simpler Solutions

no code implementations • 7 Sep 2023 • Nagabhushan Somraj, Adithyan Karanayil, Rajiv Soundararajan

The depth estimated by these simpler models is used to supervise the NeRF depth estimates.

ViP-NeRF: Visibility Prior for Sparse Input Neural Radiance Fields

1 code implementation • 28 Apr 2023 • Nagabhushan Somraj, Rajiv Soundararajan

We reformulate the NeRF to also directly output the visibility of a 3D point from a given viewpoint to reduce the training time with the visibility constraint.

Depth Estimation

Temporal View Synthesis of Dynamic Scenes through 3D Object Motion Estimation with Multi-Plane Images

1 code implementation • 19 Aug 2022 • Nagabhushan Somraj, Pranali Sancheti, Rajiv Soundararajan

The predicted object motion is then integrated with the given user or camera motion to generate the next frame.

Motion Estimation Temporal View Synthesis +1

Revealing Disocclusions in Temporal View Synthesis through Infilling Vector Prediction

1 code implementation • 17 Oct 2021 • Vijayalakshmi Kanchana, Nagabhushan Somraj, Suraj Yadwad, Rajiv Soundararajan

We consider the problem of temporal view synthesis, where the goal is to predict a future video frame from the past frames using knowledge of the depth and relative camera motion.

Temporal View Synthesis

Understanding the Perceived Quality of Video Predictions

1 code implementation • 1 May 2020 • Nagabhushan Somraj, Manoj Surya Kashi, S. P. Arun, Rajiv Soundararajan

We show that our feature design leads to state of the art quality prediction in accordance with human judgments on our IISc PVQA Database.

Representation Learning Video Prediction