Cross-domain Beauty Item Retrieval via Unsupervised Embedding Learning

  title={Cross-domain Beauty Item Retrieval via Unsupervised Embedding Learning},
  author={Zehang Lin and H. Xie and Peipei Kang and Zhenguo Yang and Wenyin Liu and Qing Li},
  journal={Proceedings of the 27th ACM International Conference on Multimedia},
  • Zehang Lin, H. Xie, +3 authors Qing Li
  • Published 2019
  • Computer Science
  • Proceedings of the 27th ACM International Conference on Multimedia
Cross-domain image retrieval is always encountering insufficient labelled data in real world. In this paper, we propose unsupervised embedding learning (UEL) for cross-domain beauty and personal care product retrieval to finetune the convolutional neural network (CNN). More specifically, UEL utilizes the non-parametric softmax to train the CNN model as instance-level classification, which reduces the influence of some inevitable problems (e.g., shape variations). In order to obtain better… Expand
Learning to Remember Beauty Products
A deep learning model for the beauty product image retrieval problem that is fully end-to-end, requires no manual feature aggregation and post-processing, and introduces a distance loss to regularize embedding vectors from the encoder to be more discriminative. Expand
Multi-Scale Generalized Attention-Based Regional Maximum Activation of Convolutions for Beauty Product Retrieval
This paper proposes a novel descriptors, named Multi-Scale Generalized Attention-Based Regional Maximum Activation of Convolutions (MS-GRMAC), which introduces multi-scale generalized attention mechanism to reduce the influence of scale variations, thus, can boost the performance of the retrieval task. Expand
Attention-driven Unsupervised Image Retrieval for Beauty Products with Visual and Textual Clues
This work proposes a search method utilizing both images and product descriptions and experimentally proved the effectiveness of the proposed method on the validation set of the Perfect-500K dataset. Expand
Multi-Feature Fusion Method Based on Salient Object Detection for Beauty Product Retrieval
A multi-feature fusion method based on salient object detection to improve retrieval performance by to extract the foreground objects of the query set by using the salient object Detection network, so as to eliminate the background interference. Expand
Learning Visual Features from Product Title for Image Retrieval
The easily accessible text information, that is, the product title, is utilized as a supervised signal to learn the features of the product image to construct a dataset for image classification and achieve the fourth position in the Grand Challenge of AI Meets Beauty in 2020 ACM Multimedia. Expand
Fashion Meets Computer Vision
A comprehensive survey of more than 200 major fashion-related works covering four main aspects for enabling intelligent fashion and highlighting promising directions for future research. Expand
Fashion Meets Computer Vision: A Survey
A comprehensive survey of more than 200 major fashion-related works covering four main aspects for enabling intelligent fashion and highlighting promising directions for future research. Expand


Regional Maximum Activations of Convolutions with Attention for Cross-domain Beauty and Personal Care Product Retrieval
A regional maximum activations of convolutions with attention (RA-MAC) descriptor to extract image features for retrieval and achieves the 2nd place in the leader board of the Grand Challenge of AI Meets Beauty in ACM Multimedia 2018. Expand
Fine-Tuning CNN Image Retrieval with No Human Annotation
It is shown that both hard-positive and hard-negative examples, selected by exploiting the geometry and the camera positions available from the 3D models, enhance the performance of particular-object retrieval. Expand
End-to-End Learning of Deep Visual Representations for Image Retrieval
This article uses a large-scale but noisy landmark dataset and develops an automatic cleaning method that produces a suitable training set for deep retrieval, and builds on the recent R-MAC descriptor, which can be interpreted as a deep and differentiable architecture, and presents improvements to enhance it. Expand
Particular object retrieval with integral max-pooling of CNN activations
This work revisits both retrieval stages, namely initial search and re-ranking, by employing the same primitive information derived from the CNN, and significantly improves existing CNN-based recognition pipeline. Expand
Aggregating Deep Convolutional Features for Image Retrieval
This paper investigates possible ways to aggregate local deep features to produce compact global descriptors for image retrieval and shows that deep features and traditional hand-engineered features have quite different distributions of pairwise similarities, hence existing aggregation methods have to be carefully re-evaluated. Expand
Beauty Product Image Retrieval Based on Multi-Feature Fusion and Feature Aggregation
VGG16 and ResNet50 are chosen to extract image features, and Crow is adopted to perform deep feature aggregation in a beauty product image retrieval method based on multi-feature fusion and feature aggregation. Expand
Unsupervised Embedding Learning via Invariant and Spreading Instance Feature
A novel instance based softmax embedding method, which directly optimizes the `real' instance features on top of the softmax function, which achieves significantly faster learning speed and higher accuracy than all existing methods. Expand
Efficient Diffusion on Region Manifolds: Recovering Small Objects with Compact CNN Representations
This work focuses on diffusion, a mechanism that captures the image manifold in the feature space through a sparse linear system solver, yielding practical query times well below one second. Expand
Visual Instance Retrieval with Deep Convolutional Networks
This paper provides an extensive study on the availability of image representations based on convolutional networks (ConvNets) for the task of visual instance retrieval and presents an efficient pipeline exploiting multi-scale schemes to extract local features by taking geometric invariance into explicit account. Expand
Unprecedented Usage of Pre-trained CNNs on Beauty Product
This paper adopts several well known deep learning models pre-trained on ImageNet, and evaluates their performance using different distance metrics to secure fourth position in the Perfect Corp Challenge. Expand