Learning visual similarity for product design with convolutional neural networks

By Sean Bell, Kavita Bala. 2015

440 citations recorded by the source corpus.

Subfields: 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 Computer Vision and Pattern Recognition

Citations in the corpus

Revisiting IM2GPS in the Deep Learning Era (2017, background): “Image retrieval using learned, deep representations is useful to a wide range of tasks such as product ranking [4, 11], sketch based image retrieval [25], face recognition [31, 26], cross-view localization [18, 37, 34] and scene retrieval [33, 2, 23, 7].”
Learning Deep Structure-Preserving Image-Text Embeddings (2015, background): “Finally, our work is related to deep similarity learning [4, 6, 7, 16, 18, 39, 47], though we are solving a cross-view, not a within-view, matching problem.”
Learning Visual Clothing Style with Heterogeneous Dyadic Co-Occurrences (2015, methodology): “Then, we show how to train a Siamese CNN to learn a feature transformation from the image space into the latent style space.”
Deep Learning for IoT Big Data and Streaming Analytics: A Survey (2017, background): “A recent shift toward product image retrieval through visual search techniques was noticed [159].”
Set2Graph: Learning Graphs From Sets (2020, methodology): “…learning Previous models suggested for learning pairwise relations in sets were mostly of the form H 2 ( X ) = m ( φ ( x i ) , φ ( x j )) (Hsu et al., 2017; Bromley et al., 1994; Simo-Serra et al., 2015; Zagoruyko & Komodakis, 2015; Bell & Bala, 2015; Ahmed et al., 2015; Vo & Hays, 2016).”
Deep Metric Learning via Lifted Structured Feature Embedding (2015, methodology): “However, the approach is not scalable with the size of the training data because it’s very costly to obtain ranking annotations in contrast to multiclass labels (i.e., product name) and because the approach is limited to ranking the data in triplet form.”
Marr Revisited: 2D-3D Alignment via Surface Normal Prediction (2016, methodology): “Inspired by [3], we seek to model style given a training set of hand-aligned similar and dissimilar CAD model-image pairs.”
Sampling Matters in Deep Embedding Learning (2017, background): “Models that transform images into rich, semantic representations lie at the heart of modern computer vision, with applications ranging from zero-shot learning [5, 41] and visual search [3, 10, 22, 29], to face recognition [6, 23, 25, 28] or fine-grained retrieval [22, 28, 29].”
Generating Liquid Simulations with Deformation-aware Neural Networks (2017, methodology): “ld of computer graphics, researchers have also employed neural networks for diverse tasks such as robust human correspondences from depth maps [Wei et al. 2015], visual similarity for product design [Bell and Bala 2015], automatic image colorization [Iizuka et al. 2016] and photo retrieval from sket”
Localizing and Orienting Street Views Using Overhead Imagery (2016, methodology): “For example, ranking can be improved by adding a classiﬁcation layer predicting category [4, 25] or attributes [14].”
Answering Visual-Relational Queries in Web-Extracted Knowledge Graphs (2017, methodology): “There is a large body of work on metric learning where the objective is to generate image embeddings such that a pairwise distance-based loss is minimized [Schroff et al., 2015, Bell and Bala, 2015, Oh Song et al., 2016, Sohn, 2016, Wang et al., 2017].”
Fully Convolutional Attention Networks for Fine-Grained Recognition (2016, background): “With the great potential in rivaling human experts, it has shown tremendous applications in real world ranging from e-commerce [6, 7] to education [8, 9].”
Deep Exemplar 2D-3D Detection by Adapting from Real to Rendered Views (2015, background): “, AlexNet [31] “pool5” features) [4], learning a transformation from CNN features to light-field descriptors for 3D shapes [35], and training a Siamese network for style retrieval [6].”
Deep Metric Learning: A Survey (2019, background): “In the beginning, some papers tend to use easy sample pairs for the Siamese network in embedding learning [29,88].”
Representation Learning for Visual-Relational Knowledge Graphs (2017, background): “…a large body of work on metric learning where the objective is to generate image embeddings such that a pairwise distance-based loss is minimized [4, 33, 39, 43, 50]. n p Veit et al. [49] propose a siamese CNN to learn a metric representation on pairs of textile products so as to learn which…”
Deep Convolutional Neural Networks for Image Classification: A Comprehensive Review (2017, methodology): “…classification–related applications that have used the contrastive loss as part of DCNN architecture include face representation (Sun, Chen, Wang, & Tang, 2014) and visual similarity for visual search (Bell & Bala, 2015), where the contrastive loss was used in combination with the softmax loss.”
A survey of deep neural network architectures and their applications (2017, background): “Recent researches have shown that in c lassification tasks, deep learning techniques can obtain promising results [9], [89].”
Visual Discovery at Pinterest (2017, background): “1 Recent work has also demonstrated the effectiveness of learning embeddings or distance functions directly from ranking labels such as relative comparisons [33] [20] and variations of pairwise comparisons [6] [38], or using Bilinear features [13] for fine-grained recognition.”