FashionAI: A Hierarchical Dataset for Fashion Understanding

Fashion Meets Computer Vision

“…There are some evaluation metrics used to assess the performance of clothing retrieval methods. The different evaluation metrics used are as follows: (1) Top-k retrieval accuracy, the ratio of queries with at least one matching item retrieved within the top-k returned results, (2) Precision@k, the ratio of items in the top-k returned results that are matched Fashion Meets Computer Vision: A Survey 72:11 Since dataset for attribute-specific fashion retrieval is lacking, this dataset rebuild three fashion dataset with attribute annotations DARN [69], FashionAI [232], DeepFashion [123] with the queries, (3) Recall@k, the ratio of matching items that are covered in the top-k returned results, (4) Normalized Discounted Cumulative Gain (NDCG) (NDCG@k), the relative orders among matching and non-matching items within the top-k returned results, and ( 5) Mean Average Precision (MAP), which measures the precision of returned results at every position in the ranked sequence of returned results across all queries. Table 2 of the Supplementary Material presents the performance comparisons of some retrieval methods reviewed in this survey.…”

Construction machine pose prediction considering historical motions and activity attributes using gated recurrent unit (GRU)

“…the forecasted keypoints and the actual values [48]. Only the keypoints included in the input frame (i.e.…”

You can Try without Visiting: A Comprehensive Survey on Virtually Try-on Outfits

“…Physical simulation working within the fashion domain focus on clothing-body interactions, and datasets can be categorized into real data and created data. Despite the rapid revolution on previous datasets which are based on 2D images like DeepFashion [72], DeepFashion2 [77] and FashionAI [116], the production of datasets basing on 3D clothing is almost rare or not sufficient for training like the digital wardrobe released by MGN [117]. In 2020, Heming et al [118] develop a comprehensive dataset named Deep Fashion3D which is richly annotated and covers a much larger variations of garment styles.…”