Multimodal Fusion of Deeply Inferred Point Clouds for 3D Scene Reconstruction Using Cross-Entropy ICP.
Watcharaphong YookwanKrisana ChinnasarnChakchai So-InParamate HorkaewPublished in: IEEE Access (2022)
Keyphrases
- point cloud
- cross entropy
- multimodal fusion
- iterative closest point
- high robustness
- audio visual
- relevance feedback
- structure from motion
- log likelihood
- icp algorithm
- multimodal interfaces
- point sets
- maximum likelihood
- gait recognition
- epipolar geometry
- laser scanner
- evaluation metrics
- language modeling
- error function
- ranking functions
- neural network
- human computer interaction
- artificial neural networks
- image processing
- stereo camera
- scoring function
- nearest neighbor
- image sequences
- information retrieval
- machine learning