#multi-modal
VisualGLM-6B
VisualGLM-6B is a multi-modal dialog language model supporting images, Chinese, and English, based on ChatGLM-6B with 7.8 billion parameters including visual capabilities from BLIP2-Qformer. The model achieves visual-linguistic interoperability and can be deployed on consumer GPUs by using quantized accuracy. It is pre-trained on 330 million captioned images, optimizing alignment across languages while adhering to open-source protocols. Limitations include image specificity and potential model hallucinations, with plans for future improvements.
MM-Interleaved
The MM-Interleaved model is a pioneer in interleaved image-text generative modeling, featuring a multi-modal feature synchronizer for high-resolution recognition. It supports tasks like visual storytelling, visual question answering, and text-to-image generation. With zero-shot and finetuning capabilities, it offers excellent performance across multiple benchmarks. Access pretrained models for versatile application adaptation.
Macaw-LLM
Explore the innovative integration of images, audio, video, and text data in this project. Utilizing leading models like CLIP, Whisper, and LLaMA, the project offers efficient alignment for multi-modal data, along with features such as one-stage instruction fine-tuning and a novel multi-modal instruction dataset. An ideal tool for investigating the prospects of multi-modal LLMs, fostering research for comprehending intricate real-world situations.
Awesome-LM-SSP
This repository emphasizes the multi-dimensional trustworthiness of large models, specifically in safety, security, and privacy. It includes a comprehensive collection of resources on multi-modal models such as vision-language and diffusion models. It serves as an essential tool for researchers by offering an extensive selection of papers and the option for users to recommend additional resources. The repository keeps users informed with recent academic advancements and categorizations, making it a critical asset for scholarly and practical pursuits.
ustore
This open-source modular database is crafted for flexibility and high performance, catering to AI and semantic search with robust ACID guarantees. It stands out with compatibility across storage backends like RocksDB and LevelDB, and handles Blobs, Documents, Graphs, and Vectors efficiently. With seamless integration via Python, C, GoLang, and Java drivers, it addresses diverse data management needs, from document storage to vector search, while enhancing capabilities with tools like Pandas and NetworkX. Remote access is streamlined through the Apache Arrow Flight interface, presenting an adaptable solution that can potentially replace multiple other databases in various AI applications.
EmbodiedScan
EmbodiedScan enhances embodied AI with a robust multi-modal 3D dataset, supporting effective visual grounding and scene interaction tasks in varied environments. The dataset offers over 5k scans, 1M ego-centric RGB-D views, and 160k categorized 3D boxes, bridging scene perception and language interaction. The Embodied Perceptron, a baseline framework, advances input processing for both structured tasks and real-world applications, with improvements such as dense semantic occupancy mapping and LVIS category compatibility.
Collaborative-Diffusion
The project introduces advancements in multi-modal face generation and editing through pre-trained uni-modal diffusion models. It allows precise generation and editing of high-quality facial images via text and segmentation mask controls, focusing on identity preservation and dynamic diffusion. Notable updates involve FreeU integration and comprehensive training pipelines. This repository is a valuable resource for researchers and developers in facial image synthesis and modification.
LaVIT
The LaVIT repository enhances language models by merging visual comprehension and generation into a cohesive framework. Highlighted at ICLR 2024, it utilizes visual tokenization to transform imagery into digestible data tokens, thus optimizing multimodal interaction. Video-LaVIT extends this capability to handle video content, providing reliable text-to-visual and video translation for diverse AI applications. The release of pre-trained weights on HuggingFace broadens its utility in tasks such as captioning and Q&A, supporting comprehensive multimodal operations within an integrated platform.
MultiModalMamba
Discover the AI model MultiModalMamba that combines Vision Transformer and Mamba, built on the Zeta framework for efficient multi-modal data processing. This model handles both text and image data seamlessly, making it suitable for a range of AI tasks. With customizable parameters and the option to return embeddings, it can be tailored for diverse needs such as transfer learning. MultiModalMamba provides a versatile and efficient AI solution for streamlining workflows.
Feedback Email: [email protected]