2 个仓库
Awesome GitHub RepositoriesMultilingual Image-Text Alignment
Training processes that map images and text across multiple languages into a shared vector space.
Distinct from Caption-Based Training: Extends caption-based training to include shared latent spaces for multilingual alignment, not just image-text association.
Explore 2 awesome GitHub repositories matching artificial intelligence & ml · Multilingual Image-Text Alignment. Refine with filters or upvote what's useful.
Awesome Multilingual Image-Text Alignment GitHub Repositories
用 AI 发现最棒的仓库。我们将通过 AI 为您搜索最匹配的仓库。
- kwai-kolors/kolors
Kwai-Kolors/Kolors
4,607在 GitHub 上查看Kolors 是一个生成式模型实现，用于根据自然语言描述和视觉参考合成逼真的图像。它利用潜在扩散模型框架来生成高保真图像，并在压缩的潜在空间内运行以提高生成效率和质量。 该系统作为一个多语言图像生成器，能够解释多种语言的文本提示词，从而产生语义准确的视觉输出。它包含一个自定义模型训练流水线，使用低秩自适应（LoRA）技术，通过少量图像集教会模型特定的主题或艺术风格。 该项目涵盖了广泛的图像合成和编辑功能，包括文生图和图生图转换。它提供了通过深度图或姿态图进行空间布局控制的工具、用于美学一致性的视觉身份注入，以及基于掩码的修复（Inpainting）以重构或修改特定图像区域。 该实现包括图像质量评估工具，根据人类偏好指标对生成图像的美学和语义质量进行评分。
Provides the ability to interpret text prompts in multiple languages to produce semantically accurate visual outputs.
Python - google-research/big_vision
google-research/big_vision
3,363在 GitHub 上查看This project is a research framework and toolkit designed for training large-scale vision transformers and multimodal language models. It provides a comprehensive suite for vision-language pretraining, enabling the development of models that map images and text into shared latent spaces. The framework is distinguished by its capabilities in high-fidelity image generation and multimodal research, utilizing normalizing flows and variational autoencoders to produce images from text prompts or class labels. It supports the development of both generative and contrastive models, allowing for a wide
Maps images and text into a shared space using captioning-based pretraining and self-supervised losses.
Jupyter Notebook
