DeepFashion2: A Versatile Benchmark for Detection, Pose Estimation, Segmentation and Re-Identification of Clothing Images

@article{Ge2019DeepFashion2AV,
  title={DeepFashion2: A Versatile Benchmark for Detection, Pose Estimation, Segmentation and Re-Identification of Clothing Images},
  author={Yuying Ge and Ruimao Zhang and Lingyun Wu and Xiaogang Wang and Xiaoou Tang and Ping Luo},
  journal={2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
  year={2019},
  pages={5332-5340}
}
  • Yuying Ge, Ruimao Zhang, +3 authors Ping Luo
  • Published 2019
  • Computer Science
  • 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
Understanding fashion images has been advanced by benchmarks with rich annotations such as DeepFashion, whose labels include clothing categories, landmarks, and consumer-commercial image pairs. [...] Key Method A strong baseline is proposed, called Match R- CNN, which builds upon Mask R-CNN to solve the above four tasks in an end-to-end manner. Extensive evaluations are conducted with different criterions in Deep- Fashion2. DeepFashion2 Dataset will be released at : https://github.com/switchablenorms…Expand
EPYNET: Efficient Pyramidal Network for Clothing Segmentation
TLDR
A new measure of dataset imbalance, motivated by the difficulty in comparing different datasets for clothing segmentation, is introduced and can be potentially useful for many real-world applications related to soft biometrics, people surveillance, image description, clothes recommendation, and others. Expand
A Strong Baseline for Fashion Retrieval with Person Re-Identification Models
TLDR
A simple baseline model for fashion retrieval is introduced, significantly outperforming previous state-of-the-art results despite a much simpler architecture and a cross-domain (cross-dataset) evaluation method is proposed to test the robustness of fashion retrieval models. Expand
A Global-Local Embedding Module for Fashion Landmark Detection
TLDR
A fashion landmark detection network with a global-local embedding module based on a non-local operation for capturing long-range dependencies and a subsequent convolution operation for adopting local neighborhood relations that outperforms the state-of-the-art methods. Expand
Multiple-Clothing Detection and Fashion Landmark Estimation Using a Single-Stage Detector
TLDR
A one-stage detector that rapidly detects multiple cloths and landmarks in fashion images that has the advantage of operating in low-power devices and a low number of parameters and low computational cost make it efficient. Expand
Fashion Landmark Detection and Category Classification for Robotics
TLDR
This work proposes training data augmentation methods such as elastic warping, and model adjustments such as rotation invariant convolutions to make the model generalize better, and demonstrates that this approach outperforms stateof-the art models with respect to clothing category classification and fashion landmark detection when tested on previously unseen datasets. Expand
KGDet: Keypoint-Guided Fashion Detection
TLDR
This work designs a simple yet effective Keypoint-Guided clothing Detector, named KGDet, which obtains superior performance on the DeepFashion2 dataset and the FLD dataset with high efficiency. Expand
Unselfie: Translating Selfies to Neutral-pose Portraits in the Wild
TLDR
A novel photographic transformation that automatically translates a selfie into a neutral-pose portrait and a new three-stage pipeline, where the user first finds a target neutral pose, inpaint the body texture, and finally refine and composite the person on the background is proposed. Expand
Transfer Learning for Pose Estimation of Illustrated Characters
TLDR
This work bridges the domain gap by efficiently transfer-learning from both domain-specific and task-specific source models and applies the resultant state-of-the-art character pose estimator to solve the novel task of pose-guided illustration retrieval. Expand
LG-VTON: Fashion Landmark Meets Image-Based Virtual Try-On
TLDR
A Landmark-Guided Virtual Try-On Network (LG-VTON) is proposed, which explicitly divides the clothes into regions using estimated landmarks, and performs a part-wise transformation using the Thin Plate Spline for each region independently. Expand
Towards Photo-Realistic Virtual Try-On by Adaptively Generating$\leftrightarrow$Preserving Image Content
TLDR
This work proposes a novel visual try-on network, namely Adaptive Content Generating and Preserving Network (ACGPN), which can generate photo-realistic images with much better perceptual quality and richer fine-details. Expand
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 21 REFERENCES
DeepFashion: Powering Robust Clothes Recognition and Retrieval with Rich Annotations
TLDR
This work introduces DeepFashion1, a large-scale clothes dataset with comprehensive annotations, and proposes a new deep model, namely FashionNet, which learns clothing features by jointly predicting clothing attributes and landmarks. Expand
Unconstrained Fashion Landmark Detection via Hierarchical Recurrent Transformer Networks
TLDR
This work presents a novel Deep LAndmark Network (DLAN), where bounding boxes and landmarks are jointly estimated and trained iteratively in an end-to-end manner, and presents a large-scale fashion landmark dataset, namely Unconstrained Landmark Database (ULD), consisting of 30K images. Expand
Fashion Landmark Detection in the Wild
TLDR
Fashion landmark is compared to clothing bounding boxes and human joints in two applications, fashion attribute prediction and clothes retrieval, showing that fashion landmark is a more discriminative representation to understand fashion images. Expand
Deep domain adaptation for describing people based on fine-grained clothing attributes
TLDR
A novel double-path deep domain adaptation network to model the data from the two domains jointly and several alignment cost layers placed inbetween the two columns ensure the consistency of the two domain features and the feasibility to predict unseen attribute categories in one of the domains. Expand
ModaNet: A Large-scale Street Fashion Dataset with Polygon Annotations
TLDR
A new database called ModaNet, a large-scale collection of images based on Paperdoll dataset, which allows to measure the performance of state-of-the-art algorithms for object detection, semantic segmentation and polygon prediction on street fashion images in detail. Expand
MVC: A Dataset for View-Invariant Clothing Retrieval and Attribute Prediction
TLDR
This work constructs a challenge clothing dataset, called Multi-View Clothing dataset, which not only has four different views for each clothing item, but also provides 264 attributes for describing clothing appearance and adopts a state-of-the-art deep learning method. Expand
Clothing Co-parsing by Joint Image Segmentation and Labeling
This paper aims at developing an integrated system of clothing co-parsing, in order to jointly parse a set of clothing images (unsegmented but annotated with tags) into semantic configurations. WeExpand
Mask R-CNN
TLDR
This work presents a conceptually simple, flexible, and general framework for object instance segmentation, which extends Faster R-CNN by adding a branch for predicting an object mask in parallel with the existing branch for bounding box recognition. Expand
Microsoft COCO: Common Objects in Context
We present a new dataset with the goal of advancing the state-of-the-art in object recognition by placing the question of object recognition in the context of the broader question of sceneExpand
Where to Buy It: Matching Street Clothing Photos in Online Shops
TLDR
Three different methods for Exact Street to Shop retrieval are developed, including two deep learning baseline methods, and a method to learn a similarity measure between the street and shop domains. Expand
...
1
2
3
...