VGA: Vision GUI Assistant - Minimizing Hallucinations through Image-Centric Fine-Tuning.
Ziyang MengYu DaiZezheng GongShaoxiong GuoMinglong TangTongquan WeiPublished in: CoRR (2024)
Keyphrases
- fine tuning
- image content
- image features
- image analysis
- image classification
- multiscale
- single image
- input image
- edge detection
- image data
- image segmentation
- template matching
- hough transform
- real time
- image representation
- feature points
- image synthesis
- test bed
- vector field
- region of interest
- image collections
- grey level
- vision system
- dynamic range
- keypoints
- high resolution
- image retrieval
- energy function
- visual perception
- graphical user interface
- low level vision
- image pixels
- lighting conditions
- spatial information
- image regions
- low level
- feature vectors
- image processing
- computer vision