RUST: Latent Neural Scene Representations from Unposed Imagery.
Mehdi S. M. SajjadiAravindh MahendranThomas KipfEtienne PotDaniel DuckworthMario LucicKlaus GreffPublished in: CVPR (2023)
Keyphrases
- d scene
- single image
- network architecture
- aerial video
- neural network
- video sequences
- distributed representations
- three dimensional
- computer vision
- multiple images
- scene analysis
- visual motion
- complex scenes
- latent variables
- man made
- scene understanding
- observed scene
- imaging sensors
- scene images
- satellite imagery
- object models
- image data
- associative memory
- higher level
- image sequences
- real scenes
- scene structure
- high resolution
- image processing
- internal representations
- input image
- multi view
- scene reconstruction
- video data
- neural model
- scene classification
- dynamic scenes
- geometric constraints