Upsampling
رفع الدقة
زيادة الدقة المكانية (الأبعاد) لخريطة السمات. يمكن أن تتم بالإقحام الثنائي الخطي (يدوي) أو بالتفافات منقولة (مُتعلَّمة). عملية أساسية في فاك الترميز لاستعادة الحجم الأصلي للصورة.
Increasing the spatial resolution (dimensions) of a feature map. Can be done via bilinear interpolation (hand-designed) or transposed convolutions (learned). A core operation in decoders to recover the original image size.
Also translated asإعادة التكبير، الترقية المكانية
First appears in this corpus in: Fully Convolutional Networks for Semantic Segmentation (2015)
Appears in these papers
- DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs2017in the sky ✦
- Diffusion Models Beat GANs on Image Synthesis2021in the sky ✦
- Diffusion Models Beat GANs on Image Synthesis2021in the sky ✦
- Fully Convolutional Networks for Semantic Segmentation2015in the sky ✦
- GLIDE: Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion Models2021in the sky ✦
- Jukebox: A Generative Model for Music2020in the sky ✦
- Jukebox: A Generative Model for Music2020in the sky ✦
- Progressive Growing of GANs for Improved Quality, Stability, and Variation2018in the sky ✦
- RAFT: Recurrent All-Pairs Field Transforms for Optical Flow2020in the sky ✦
- RAFT: Recurrent All-Pairs Field Transforms for Optical Flow2020in the sky ✦
- Analyzing and Improving the Image Quality of StyleGAN2020in the sky ✦
- YOLOv3: An Incremental Improvement2018in the sky ✦