{"id":489125,"date":"2018-06-05T09:23:14","date_gmt":"2018-06-05T16:23:14","guid":{"rendered":"https:\/\/www.microsoft.com\/en-us\/research\/?post_type=msr-blog-post&p=489125"},"modified":"2018-06-05T09:23:14","modified_gmt":"2018-06-05T16:23:14","slug":"netizen-style-commenting-fashion-photos-autonomous-diverse-cognitive","status":"publish","type":"msr-blog-post","link":"https:\/\/www.microsoft.com\/en-us\/research\/articles\/netizen-style-commenting-fashion-photos-autonomous-diverse-cognitive\/","title":{"rendered":"Netizen-style commenting for fashion photos: autonomous, diverse, and cognitive"},"content":{"rendered":"
The advance of deep neural networks has brought huge advances in image captioning. However, current work is deficient in several ways. It simply generates \u201cvanilla\u201d sentences, which describe the shallow appearance of things (e.g., color, types) in the photo and typically doesn\u2019t create a caption with engaging information about context or their intentions, in a way that a human would.<\/p>\n
Recently, Professor Winston Hsu from National Taiwan University collaborated with researchers at Microsoft Research Asia (MSRA) to address this challenge in social media photo commenting for user-contributed fashion photos.<\/p>\n