Janey[paper reading] Segment AnythingMain contribution: 1 Billion masks & 11M images; a promptable model which is easy to transfer to zero-shot.Apr 20, 2023Apr 20, 2023
Janey[Paper reading] Segmentation: Masked-attention Mask Transformer for Universal Image SegmentationMain goal: Anew architecture capable of addressing any image segmentation task (panoptic, instance or semantic).Apr 19, 2023Apr 19, 2023
Janey[Paper reading] DERT: Object DetectionI still have some questions on this paper! If there is anyone interested in, please help me outΒ :)Apr 17, 2023Apr 17, 2023
Janey[Paper reading]Masked AutoencodersPaper1: Masked Autoencoders Are Scalable Vision LearnersApr 4, 2023Apr 4, 2023