Another clue highlighting the value of NeRF is in the name, ‘radiance ’. This means that given a set of images (even a sparse incomplete set) taken for a specific scene, a NeRF can be trained to generate views from viewing angles it has not seen before. While #generativeai ( #gpt4, #bard, #stablediffusion #midjourney) took centre stage, one of the most important technologies to emerge from this wave of innovation, NeRFs (or Neural Radiance Fields) continued to quietly develop in the background surpassing many of the perceived limitations.Ī NeRF uses #deeplearning to infer a continuous 3D representation of a scene from a set of 2D images.
With all of the excitement of Q1 I forgot about NeRFs, but the NeRF community didn't stop developing.
0 kommentar(er)
