Deep Spatial-Semantic Attention for Fine-Grained Sketch-Based Image Retrieval

By Jifei Song, Qian Yu, Yi-Zhe Song, Tao A. Xiang, Timothy M. Hospedales. 2017

256 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

Doodle It Yourself: Class Incremental Learning by Drawing a Few Sketches (2022, background): “Apart from the widely studied SBIR [11, 67, 12, 41, 46, 34, 53, 6, 2], sketch has also been employed in a variety of vision understanding tasks, including segmentation [22], video synthesis [28], representation learning [60, 3], object localisation [55], image-inpainting [62], 3D shape retrieval…”
A Decade Survey of Content Based Image Retrieval Using Deep Learning (2020, methodology): “loss [72], higher-order learnable energy function (HOLEF) based loss [180], ranking loss [116], scaling and shifting based adaptive loss function [129], holographic composition layer based loss”
StyleMeUp: Towards Style-Agnostic Sketch-Based Image Retrieval (2021, methodology): “Starting with deformable-part models [27], various deep approaches have surfaced with the advent of new FG-SBIR datasets [49, 60, 20].”
Sketch3T: Test-Time Training for Zero-Shot SBIR (2022, methodology): “Aided by new datasets [54, 66], FG-SBIR flourished with the introduction of triplet-ranking models [66], learning a joint sketch-photo manifold.”
More Photos are All You Need: Semi-Supervised Learning for Fine-Grained Sketch Based Image Retrieval (2021, methodology): “7K for QMUL-ShoeV2 [48, 39], and 2K for QMUL-ChairV2 [39]), while photodatasets [32] are available in millions.”
Learning to Sketch with Shortcut Cycle Consistency (2018, unknown): “The relevant research field has flourished recently [22, 45, 36, 32, 21, 28, 41, 35] due to its huge potential for commercial applications.”
Semantically Tied Paired Cycle Consistency for Zero-Shot Sketch-Based Image Retrieval (2019, background): “[47] without the shortcut connection for extracting only the informative regions from sketch and image.”
Deep Learning for Free-Hand Sketch: A Survey (2020, background): “(v) More sketch-based applications have appeared, i . e ., on-line sketch game QuickDraw 2 [16], sketch-based commodity search engine 3 [5], [20].”
Adaptive Triplet Model for Fine-Grained Visual Categorization (2018, background): “It is widely used in many fields, such as face recognition [30], image retrial [31], [32] and person re-identification [33].”
Partially Does It: Towards Scene-Level FG-SBIR with Partial Input (2022, methodology): “HOLEF [64] extends Triplet-SN by adding spatial attention along with higher order ranking loss.”
Fine-Grained Image Analysis With Deep Learning: A Survey (2021, background): “[54] introduced an attention module and encoded the”
Cross-domain Generative Learning for Fine-Grained Sketch-Based Image Retrieval (2017, background): “It has received increasing research attention recently [12, 14, 19, 22, 23, 27], due to its potential in commercial applications such as searching online product catalogues for shoes, furniture, and handbags by ﬁnger-sketching on a touchscreen device.”
Sketching without Worrying: Noise-Tolerant Sketch-Based Image Retrieval (2022, methodology): “Initially starting with graph-matching models [33], FG-SBIR research gained traction with the advent of various deeplearning based approaches [56, 46, 7, 3].”
Generalising Fine-Grained Sketch-Based Image Retrieval (2019, methodology): “∗Equal Contribution Recent FG-SBIR methods [24, 36, 28, 22] address this issue by learning a deep network embedding of sketch and photo that makes them directly comparable.”
Vectorization and Rasterization: Self-Supervised Learning for Sketch and Handwriting (2021, background): “While rasterized sketch images are usually claimed to be better for driving fine-grained retrieval [64, 58], they fail to model the varying level of abstraction [57] in the sketch generative process.”
Sketch Less for More: On-the-Fly Fine-Grained Sketch-Based Image Retrieval (2020, unknown): “This paradigm was subsequently improved through hybrid generative-discriminative cross-domain image generation [30]; and providing an attention mechanism for finegrained details as well as more sophisticated triplet losses [41].”