Learning Discrete Hashing Towards Efficient Fashion Recommendation

Abstract

In our daily life, how to match clothing well is always a troublesome problem especially when we are shopping online to select a pair of matched pieces of clothing from tens of thousands available selections. To help common customers overcome selection issues, recent studies in the recommender system area have started to infer the fashion matching results automatically. The traditional fashion recommendation is normally achieved by considering visual similarity of clothing items or/and item co-purchase history from existing shopping transactions. Due to the high complexity of visual features and the lack of historical item purchase records, most of the existing work is unlikely to make an efficient and accurate recommendation. To address the problem, in this paper, we propose a new model called Discrete Supervised Fashion Coordinates Hashing. Its main objective is to learn meaningful yet compact high-level features of clothing items, which are represented as binary hash codes. In detail, this learning process is supervised by a clothing matching matrix, which is initially constructed based on limited known matching pairs and subsequently on the self-augmented ones. The proposed model jointly learns the intrinsic matching patterns from the matching matrix and the binary representations from the clothing items' images, where the visual feature of each clothing item is discretized into a fixed-length binary vector. The binary representation learning significantly reduces the memory cost and accelerates the recommendation speed. The experiments compared with several state-of-theart approaches have evidenced the superior performance of the proposed approach on efficient fashion recommendation.

A Review of Modern Fashion Recommender Systems

“…In Table 3, we provide a classification of data/computational features used for the design of fashion RS by authors over the years. ✓ ✓ ✓ ✓ Hu et al [65] ✓ ✓ Cardoso et al [21] ✓ ✓ ✓ Jagadeesh et al [71] ✓ ✓ ✓ Liu et al [94] ✓ ✓ Zhang et al [163] ✓ ✓ Qian et al [115] ✓ ✓ ✓ ✓ ✓ ✓ Zhao et al [164] ✓ ✓ ✓ McAuley et al [107] ✓ ✓ Pan et al [111] ✓ Sun et al [131] ✓ ✓ ✓ Tangseng and Okatani [133] ✓ ✓ ✓ Sridevi et al [130] ✓ Brand and Gross [20] ✓ ✓ Xing et al [ [120]. Matrix factorization (MF), or BPR-MF, is one of the most promising approaches to building CF.…”

Snapshot boosting: a fast ensemble framework for deep neural networks

“…Deep neural network (DNN) has been widely used and achieved great performance on many practical applications [1][2][3]. Usually, we train a single DNN until convergence, but the generalization ability of a single model is limited [4].…”

System for styling apparels that match with occasion wearing via E-commerce channel using machine learning

“…It is labeling text input by machine. Then, this method uses K-Nearest Neighbor similar items to recommend matching or similar apparel (L. Liu, Du, Zhu, Shen, & Huang, 2018). This recommendation system is suitable for classic apparel stores that focus on their repeated customers, but not for the fashion product.…”