paper-with-me

홈 › Papers

Arbitrary-Scale Image Synthesis

2022-04-05 · CVPR 2022 1 · Evangelos Ntavelis, Mohamad Shahbazi, Iason Kastanis, Radu Timofte, Martin Danelljan, Luc van Gool

Positional encodings have enabled recent works to train a single adversarial network that can generate images of different scales. However, these approaches are either limited to a set of discrete scales or struggle to maintain good perceptual quality at the scales for which the model is not trained explicitly. We propose the design of scale-consistent positional encodings invariant to our generator's layers transformations. This enables the generation of arbitrary-scale images even at scales unseen during training. Moreover, we incorporate novel inter-scale augmentations into our pipeline and partial generation training to facilitate the synthesis of consistent images at arbitrary scales. Lastly, we show competitive results for a continuum of scales on various commonly used datasets for image synthesis.

📄 PDF Abstract BibTeX arXiv:2204.02273

Code (1)

vglsd/scaleparty 공식 구현 pytorch

Tasks

Image Generation

Similar Papers 제목 키워드 기반

InfiniCity: Infinite-Scale City Synthesis

2023-01-23 · ICCV 2023 1 · Chieh Hubert Lin, Hsin-Ying Lee, Willi Menapace, Menglei Chai 외

Toward infinite-scale 3D city synthesis, we propose a novel framework, InfiniCity, which constructs and renders an unconstrainedly large and 3D-grounded environment from random noises. InfiniCity decomposes the seemingly…

Image GenerationNeural Rendering

ASSR-NeRF: Arbitrary-Scale Super-Resolution on Voxel Grid for High-Quality Radiance Fields Reconstruction

2024-06-28 · Ding-Jiun Huang, Zi-Ting Chou, Yu-Chiang Frank Wang, Cheng Sun

NeRF-based methods reconstruct 3D scenes by building a radiance field with implicit or explicit representations. While NeRF-based methods can perform novel view synthesis (NVS) at arbitrary scale, the performance in high…

Image Super-ResolutionNeRFNovel View SynthesisSuper-Resolution

Example-Guided Image Synthesis across Arbitrary Scenes using Masked Spatial-Channel Attention and Self-Supervision

2020-04-18 · Haitian Zheng, Haofu Liao, Lele Chen, Wei Xiong 외

Example-guided image synthesis has recently been attempted to synthesize an image from a semantic label map and an exemplary image. In the task, the additional exemplar image provides the style guidance that controls the…

Image GenerationSpatial Interpolation

BlendGAN: Implicitly GAN Blending for Arbitrary Stylized Face Generation

2021-10-22 · NeurIPS 2021 12 · Mingcong Liu, Qiang Li, Zekui Qin, Guoxin Zhang 외

Generative Adversarial Networks (GANs) have made a dramatic leap in high-fidelity image synthesis and stylized face generation. Recently, a layer-swapping mechanism has been developed to improve the stylization performan…

DiversityFace GenerationImage Generation

Squeeze, Recover and Relabel: Dataset Condensation at ImageNet Scale From A New Perspective

2023-06-22 · NeurIPS 2023 11 · Zeyuan Yin, Eric Xing, Zhiqiang Shen

We present a new dataset condensation framework termed Squeeze, Recover and Relabel (SRe$^2$L) that decouples the bilevel optimization of model and synthetic data during training, to handle varying scales of datasets, mo…

Bilevel OptimizationDataset CondensationDataset Distillation