Deep domain adaptation for describing people based on fine-grained clothing attributes

By Qiang Chen, Junshi Huang, Rogério Schmidt Feris, Lisa Gottesfeld Brown, Jian Dong, Shuicheng Yan. 2015

264 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

Citations in the corpus

Diagnosing Gender Bias in Image Recognition Systems (2020, background): “Gathering tens of thousands or more images, all with labels describing their content, remains both costly and time consuming (Chen et al. 2015).”
Domain-Adversarial Training of Neural Networks (2015, background): “Another approach to deep domain adaptation, which is arguably more diﬀerent from ours, has been developed in parallel by Chen et al. (2015).”
Unified Deep Supervised Domain Adaptation and Generalization (2017, background): “Among algorithms for subspace learning, Siamese networks [11] work well for different tasks [14, 55, 35, 63, 9].”
Pixel-Level Domain Transfer (2016, background): “[13] also has presented a similar fashion dataset dealing with two domains.”
SpotTune: Transfer Learning Through Adaptive Fine-Tuning (2018, background): “e, [3, 4, 6, 3]) with downsampling layers in between.”
Fully-Adaptive Feature Sharing in Multi-Task Networks with Applications in Person Attribute Classification (2016, background): “State-of-the-art methods for person attribute prediction are based on deep convolutional neural networks [43, 24, 5, 48].”
DeepFashion2: A Versatile Benchmark for Detection, Pose Estimation, Segmentation and Re-Identification of Clothing Images (2019, background): “With the development of fashion datasets [20, 5, 7, 3, 14, 12, 21, 1], significant progresses have been achieved in this area [2, 19, 17, 18, 9, 8].”
Deep Attributes Driven Multi-Camera Person Re-identification (2016, methodology): “Currently, many studies have applied deep learning to attributes learning [36,37].”
DeepFashion: Powering Robust Clothes Recognition and Retrieval with Rich Annotations (2016, background): “Recently, extensive research efforts have been devoted to clothes classification [11, 1, 29], attribute prediction [3, 13, 4, 24], and clothing item retrieval [17, 6, 10, 27, 15], because of their potential values to the industry.”
Beyond Sharing Weights for Deep Domain Adaptation (2016, methodology): “The former essentially corresponds to the method proposed in [54], where the authors suggested using a two-stream architecture for domain adaptation, allowing the weights of the corresponding layers in both streams to vary independently of each other.”
Cross-Domain Image Retrieval with a Dual Attribute-Aware Ranking Network (2015, methodology): “Similar to Chen et al [5], we process the region proposals by a NIN model.”
Learning Attributes Equals Multi-Source Domain Generalization (2016, background): “Some recent work has applied deep models to attribute detection [11, 83, 47, 16].”
Fashion Landmark Detection in the Wild (2016, background): “To better capture discriminative information in fashion items, previous works have explored the usage of full image [2], general object proposals [2], bounding boxes [9,17] and even masks [22,18,23,24].”
VisDA: The Visual Domain Adaptation Challenge (2017, methodology): “The validation dataset for the classiﬁcation track is built from the Microsoft COCO [25] Training and Validation splits.”
Fashion IQ: A New Dataset Towards Retrieving Images by Natural Language Feedback (2021, methodology): “Visual attributes, including color, shape, and texture, have been successfully used to model clothing images [25, 22, 23, 2, 73, 7, 40].”
Imbalanced Deep Learning by Minority Class Incremental Rectification (2018, methodology): “To discriminate subtle classes from multi-labelled images at large scale, standard learning algorithms require a vast quantity of class balanced training data for all labels [11,18].”