Abstract: Recently deep learning methods have achieved a great success in image inpainting problem. However, reconstructing continuities of complex structures with non-stationary textures remains a challenging task for computer vision. In this paper, a novel approach to image inpainting problem is presented, which adapts exemplar-based methods for deep convolutional neural networks. The concept of onion convolution is introduced with the purpose of preserving feature continuities and semantic coherence. Similar to recent approaches, our onion convolution is able to capture long-range spatial correlations. In general, the implementation of modules with such ability in low-level features leads to impractically high latency and complexity. To address this limitations, the onion convolution suggests an efficient implementation. As qualitative and quantitative comparisons show, our method with onion convolutions outperforms state-of-the-art methods by producing more realistic, visually plausible and semantically coherent results.

SlidesLive

Similar Papers

Hyperparameter-Free Out-of-Distribution Detection Using Cosine Similarity
Engkarat Techapanurak (Tohoku University)*, Masanori Suganuma (RIKEN AIP / Tohoku University), Takayuki Okatani (Tohoku University/RIKEN AIP)
Discrete Spatial Importance-Based Deep Weighted Hashing
Yang Shi (Shandong University), Xiushan Nie (Shandong Jianzhu University)*, Quan Zhou (Shandong University), Xiaoming Xi (Shandong Jianzhu University ), Yilong Yin (Shandong University)
CLASS: Cross-Level Attention and Supervision for Salient Objects Detection
Lv Tang (Nanjing University)*, Bo Li (Nanjing University)