Attentive Fashion Grammar Network for Fashion Landmark Detection and Clothing Category Classification

Yang, Yi; Xu, Yuanlu; Shen, Jianbing; Zhu, Song‐Chun

doi:10.1109/cvpr.2018.00449

Cited by 238 publications

(174 citation statements)

References 35 publications

Supporting

Mentioning

173

Contrasting

Unclassified

Order By: Relevance

“…LGR achieves an obvious improvement over the two large datasets compared with PyraNet [32], FashionNet [36], DFA [19], DLAN [31] and BCRNNs [27]. Note that PyraNet is human pose estimation model with two stages.…”

Section: Comparison With the State-of-the-artsmentioning

confidence: 96%

“…FFLD is our contributed fine-grained fashion landmark dataset, which contains 200k images annotated with at most 32 key-points and bounding boxes for 13 clothes categories. Following [27], 209,222 fashion images are used for training; 40, 000 images are used for validation and remaining 40, 000 images are for testing in DeepFashion. Following the protocol in FLD [19], 83, 033 images and 19, 992 fashion images are used for training and validating, 19, 991 images are used for testing.…”

Section: Experimental Settingsmentioning

confidence: 99%

“…Fashion landmark detection that targets at localizing keypoints at the functional regions of clothes (e.g. collar, waistline), has attracted research attention and lots of demands driven by the boom of electronic commerce, such as clothing retrieval [36,30,5,16], clothes generation [10,21] and fashion image classification [36,27,8]. To support these comprehensive high-level applications, the landmark detectors need to effectively deal with arbitrary clothing appearances, diverse clothes layouts and styles, multiply human arXiv:1910.01923v1 [cs.CV] 4 Oct 2019 pose, different lights and background clutters.…”

Section: Introductionmentioning

confidence: 99%

“…For example, Yang et al [33] introduced to incorporate domain knowledge for explicitly facilitating features for better localization. Wang et al [27] proposed a grammar model in fashion visual understanding and used a bidirectional recurrent neural network for message passing. Fashion landmarks naturally lie in an underlying hierarchical structure which includes different levels of semantic nodes (e.g.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Layout-Graph Reasoning for Fashion Landmark Detection

Yu¹,

Liang²,

Gong³

et al. 2019

2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

View full text Add to dashboard Cite

Detecting dense landmarks for diverse clothes, as a fundamental technique for clothes analysis, has attracted increasing research attention due to its huge application potential. However, due to the lack of modeling underlying semantic layout constraints among landmarks, prior works often detect ambiguous and structure-inconsistent landmarks of multiple overlapped clothes in one person. In this paper, we propose to seamlessly enforce structural layout relationships among landmarks on the intermediate representations via multiple stacked layout-graph reasoning layers. We define the layout-graph as a hierarchical structure including a root node, body-part nodes (e.g. upper body, lower body), coarse clothes-part nodes (e.g. collar, sleeve) and leaf landmark nodes (e.g. left-collar, rightcollar). Each Layout-Graph Reasoning(LGR) layer aims to map feature representations into structural graph nodes via a Map-to-Node module, performs reasoning over structural graph nodes to achieve global layout coherency via a layout-graph reasoning module, and then maps graph nodes back to enhance feature representations via a Nodeto-Map module. The layout-graph reasoning module integrates a graph clustering operation to generate representations of intermediate nodes (bottom-up inference) and then a graph deconvolution operation (top-down inference) over the whole graph. Extensive experiments on two public fashion landmark datasets demonstrate the superiority of our model. Furthermore, to advance the fine-grained fashion landmark research for supporting more comprehensive clothes generation and attribute recognition, we contribute the first Fine-grained Fashion Landmark Dataset (FFLD) containing 200k images annotated with at most 32 keypoints for 13 clothes types.

show abstract

Section: Comparison With the State-of-the-artsmentioning

confidence: 96%

Section: Experimental Settingsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Layout-Graph Reasoning for Fashion Landmark Detection

Yu¹,

Liang²,

Gong³

et al. 2019

2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

View full text Add to dashboard Cite

show abstract

“…They mimic the human cognitive attention mechanism, which selectively focuses on the most visually informative parts of a scene. They were first explored in neural machine translation [2], and later proved effective in various natural language processing and computer vision tasks, such as image captioning [50], question answering [57], scene recognition [4,41], fashion analysis [43], etc. In the above studies, an attention mechanism is learned in a goal-driven, end-to-end manner, allowing the network to concentrate on the most task-relevant parts of the inputs.…”

Section: Related Workmentioning

confidence: 99%

Human-Aware Motion Deblurring

Shen

Yang

et al. 2019

2019 IEEE/CVF International Conference on Computer Vision (ICCV)

Self Cite

240

181

View full text Add to dashboard Cite

This paper proposes a human-aware deblurring model that disentangles the motion blur between foreground (FG) humans and background (BG). The proposed model is based on a triple-branch encoder-decoder architecture. The first two branches are learned for sharpening FG humans and BG details, respectively; while the third one produces global, harmonious results by comprehensively fusing multi-scale deblurring information from the two domains. The proposed model is further endowed with a supervised, human-aware attention mechanism in an end-toend fashion. It learns a soft mask that encodes FG human information and explicitly drives the FG/BG decoderbranches to focus on their specific domains. To further benefit the research towards Human-aware Image Deblurring, we introduce a large-scale dataset, named HIDE, which consists of 8,422 blurry and sharp image pairs with 65,784 densely annotated FG human bounding boxes. HIDE is specifically built to span a broad range of scenes, human object sizes, motion patterns, and background complexities. Extensive experiments on public benchmarks and our dataset demonstrate that our model performs favorably against the state-of-the-art motion deblurring methods, especially in capturing semantic details.

show abstract

Clothing attribute recognition via a holistic relation network

Zhou

et al. 2022

Int J of Intelligent Sys

View full text Add to dashboard Cite

Clothing attribute prediction is a fundamental image classification task in the field of computer vision. Motivated by the human recognition system, we investigate the task relationship and spatial importance where people usually utilize these useful clues to assist in recognizing clothing attributes. In this paper, we propose a novel Holistic Relation Network (HRNet) for clothing attribute recognition, considering the fusion of multiple relations, including spatial and spatial relation via a spatial relation module, spatial and task relation via a task attention module, and task and task relation via a graph context reasoning module. Specifically, we first use the backbone network to extract features from the input image, two types of attention models followed will further learn the features, then the graph context reasoning module will be used to further enhance the features, and finally, a classifier exploited to classify the clothing attributes with the learned representation information. Without using manual image feature filtering methods, this paper aims to achieve clothing attribute recognition by deeply exploring the relationships among different clothing attribute recognition tasks. In this paper, we use double‐branches of the attention model to model the relevance of spatial context information and learn more discriminative feature representations from multitask features for clothing attribute prediction. Derived from the prior knowledge learned from the two above‐mentioned attention models, we further propose a graph‐relation model constructing relationships among different clothing attribute tasks by integrating the spatial association relationships among multitask. The proposed HRNet only uses image‐level annotation but it owns a good ability for obtaining distinguishing feature representations. We obtain state‐of‐the‐art performance, which is demonstrated by extensive experiments on three mainstream benchmarks, for example, woman clothing data set, man clothing data set, and shop‐domain clothing data set.

show abstract

Attentive Fashion Grammar Network for Fashion Landmark Detection and Clothing Category Classification

Cited by 238 publications

References 35 publications

Layout-Graph Reasoning for Fashion Landmark Detection

Layout-Graph Reasoning for Fashion Landmark Detection

Human-Aware Motion Deblurring

Clothing attribute recognition via a holistic relation network

Contact Info

Product

Resources

About