Shengcao1006/RAS-13B-General
15B
•
Updated
•
8
Datasets and models for "Refer to Any Segmentation Mask Group With Vision-Language Prompts"
Note Checkpoint of RAS-13B trained on MaskGroups-2M
Note Dataset for pretraining RAS mask projector
Note Dataset for training RAS
Note Datasets for RES and GRES training and evaluation, converted from RefCOCO, RefCOCO+, RefCOCOg, and gRefCOCO in RAS format
Note Dataset for ORES evaluation