NeX: Real-time View Synthesis with Neural Basis Expansion

Suttisak Wizadwongsa*
Pakkapon Phongthawee*
Jiraphon Yenphraphai*
VISTEC - Vidyasirimedhi Institute of Science and Technology
Rayong, Thailand

CVPR 2021 (Oral, Best paper candidate)

*equal contribution


We present NeX, a new approach to novel view synthesis based on enhancements of multiplane image (MPI) that can reproduce NeXt-level view-dependent effects---in real time. Unlike traditional MPI that uses a set of simple RGBα planes, our technique models view-dependent effects by instead parameterizing each pixel as a linear combination of basis functions learned from a neural network. Moreover, we propose a hybrid implicit-explicit modeling strategy that improves upon fine detail and produces state-of-the-art results. Our method is evaluated on benchmark forward-facing datasets as well as our newly-introduced dataset designed to test the limit of view-dependent modeling with significantly more challenging effects such as the rainbow reflections on a CD. Our method achieves the best overall scores across all major metrics on these datasets with more than 1000× faster rendering time than the state of the art.

Video renderings

More results

Real-time demos

*A high-end GPU recommended for High and VR modes.

Rendering speed

This table shows the average FPS for the scenes in Shiny and Real Forward-Facing datasets. The rendering was done in real time using Google Chrome’s WebGL on a machine with an NVIDIA RTX3090. (Profiling done without Vertical Synchronization (Vsync) to go beyond the monitor’s frame rate)

Shiny dataset
CD Crest Food Gaint Lab Pasta Seasoning Tools
431 601 389 398 394 348 338 335
Real forward-facing dataset
Fern Flower Fortress Horns Leaves Orchids Room Trex
393 498 289 554 628 471 435 325
Average: 420 FPS


    author = {Wizadwongsa, Suttisak and Phongthawee, Pakkapon and Yenphraphai, Jiraphon and Suwajanakorn, Supasorn},
    title = {NeX: Real-time View Synthesis with Neural Basis Expansion},
    booktitle = {IEEE Conference on Computer Vision and Pattern Recognition (CVPR)}, 
    year = {2021},