Texture provides an important cue for many computer vision applications, and texture image classification has been an active research area over the past years. Recently, deep learning techniques using convolutional neural networks(CNN) have emerged as the state-of-the-art: CNN-based features provide...
A Sengur,I Turkoglu,MC Ince - 《Expert Systems with Applications》 被引量: 171发表: 2007年 Using Filter Banks in Convolutional Neural Networks for Texture Classification Deep learning has established many new state of the art solutions in the last decade in areas such as object, scene and spe...
4 and evaluate and conclude our method in Sects. 5 and 6. 2 Related work 2.1 Example-based stylization control Example-based methods for stylization, particularly NST as introduced by Gatys et al. [4], employ deep neural networks to extract and apply stylistic features from a reference image...
However, these volume-based methods do not (explicitly) reason about the object's surface and entangle both geometry and appearance in a volume-encoding neural network. This does not allow for easy editing—as is possi- ble with a texture mapped mesh—and significantl...
Our final results show that an SR network equipped with SFT can generate more realistic and visually pleasing textures in comparison to state-of-the-art SRGAN and EnhanceNet. 展开 关键词: Semantics Image segmentation Image resolution Modulation Visualization Training Transforms ...
This paper presents a novel semi-automatic method for lung segmentation in thoracic CT datasets. The fully three-dimensional algorithm is based on a level set representation of an active surface and integrates texture features to improve its robustness.
Next3D: Generative Neural Texture Rasterization for 3D-Aware Head Avatars Jingxiang Sun,Xuan Wang,Lizhen Wang,Xiaoyu Li,Yong Zhang,Hongwen Zhang,Yebin Liu Project|Paper|Twitter Abstract:3D-aware generative adversarial networks (GANs) synthesize high-fidelity and multi-view-consistent facial images usin...
Latterly, low light image restoration is transformed as deep iterative residual compensation, which consecutively lightens and darkens the degraded image [12]. Although previous methods can brighten low-light images with various techniques from different perspectives, these methods only consider image’s...
of Words (BoW)-based, Convolutional Neural Network (CNN)-based, and attribute-based. The BoW-based methods are organized according to their key components. The CNN-based methods are categorized into one of pretrained CNN models, finetuned CNN models, or handcrafted deep convolutional networks. ...
deep transfer learning. We use discriminative fine-tuning with one-cycle-policy and apply structure-preserving colour normalization to boost our results. We also provide visual explanations of the deep neural network’s decision on texture classification. With achieving state-of-the-art test accuracy ...