Author Archives: calvinadm

Technical report on weakly supervised object localization using things and stuff transfer

Our software release for segmentation transfer (CVPR 2012, ECCV 2012) is publicly available again.

Release of COCO-Stuff dataset

Paper on Visually Grounded Meaning Representations accepted to PAMI 2016.

Software and Dataset release for estimating the difficulty of visual search in an image.

Video data from CVPR 2010 are publicly available again.

Software release for Region-based Semantic Segmentation with End-to-End Training

Technical report on learning semantic part-based models from Google Images

Technical report on semantic part emergence in CNNs

Paper on training of object detectors for mean average precision accepted to ACCV 2016