work EmbodiedMAE Multi-modal Masked Autoencoder for 3D Plant Phenotyping Lighting-Robust Instance Segmentation SAM extended with a Lighting Convolutional Attention module FastTextSpotter High-Efficiency Transformer for Multilingual Scene Text Spotting FASTER Font-Agnostic Scene Text Editing and Rendering Framework Tricho-Vision Computer vision for trichotaxonomy and wildlife conservation Soft-set MSER Text System Occluded scene text detection, recognition, and prediction
