Abstract:Sketch recognition remains a significant challenge due to the limited training data and the substantial intra-class variance of freehand sketches for the same object. Conventional methods for this task often rely on the availability of the temporal order of sketch strokes, additional cues acquired from different modalities and supervised augmentation of sketch datasets with real images, which also limit the applicability and feasibility of these methods in real scenarios. In this paper, we propose a novel sketch-specific data augmentation (SSDA) method that leverages the quantity and quality of the sketches automatically. From the aspect of quantity, we introduce a Bezier pivot based deformation (BPD) strategy to enrich the training data. Towards quality improvement, we present a mean stroke reconstruction (MSR) approach to generate a set of novel types of sketches with smaller intra-class variances. Both of these solutions are unrestricted from any multi-source data and temporal cues of sketches. Furthermore, we show that some recent deep convolutional neural network models that are trained on generic classes of real images can be better choices than most of the elaborate architectures that are designed explicitly for sketch recognition. As SSDA can be integrated with any convolutional neural networks, it has a distinct advantage over the existing methods. Our extensive experimental evaluations demonstrate that the proposed method achieves the state-of-the-art results (84.27%) on the TU-Berlin dataset, outperforming the human performance by a remarkable 11.17% increase. Finally, more experiments show the practical value of our approach for the task of sketch-based image retrieval.

A Hybrid Convolutional Neural Network for Sketch Recognition.

Sketch-R2CNN: an Attentive Network for Vector Sketch Recognition

<i>Sketch-R2CNN</i>: An RNN-Rasterization-CNN Architecture for Vector Sketch Recognition

Sketchformer++: A Hierarchical Transformer Architecture for Vector Sketch Representation

SceneSketcher-v2: Fine-Grained Scene-Level Sketch-Based Image Retrieval Using Adaptive GCNs

Mural Sketch Generation via Style-aware Convolutional Neural Network.

Deep Neural Networks for Free-Hand Sketch Recognition.

Mixed attention dense network for sketch classification

Sequential Dual Deep Learning with Shape and Texture Features for Sketch Recognition

A Multi-Layer Deep Fusion Convolutional Neural Network for Sketch Based Image Retrieval

Deep CNN-based Features for Hand-Drawn Sketch Recognition Via Transfer Learning Approach

Entropy Information‐based Heterogeneous Deep Selective Fused Features Using Deep Convolutional Neural Network for Sketch Recognition

Sketchpointnet: A Compact Network for Robust Sketch Recognition

On-Line Sketch Recognition Using Direction Feature

A hierarchical residual network with compact triplet-center loss for sketch recognition

Application of Deep Convolutional Features in Sketch Works Classification and Evaluation

Sketch-R2CNN : An RNN-Rasterization-CNN Architecture for Vector Sketch Recognition

Sketch-Specific Data Augmentation for Freehand Sketch Recognition

End-to-End Photo-Sketch Generation via Fully Convolutional Representation Learning

On Learning Semantic Representations for Large-Scale Abstract Sketches

SketchGAN: Joint Sketch Completion and Recognition with Generative Adversarial Network.