Gradio

🆕 🤩 Negative Token Merging Performs adversarial guidance directly using images instead of text (i.e. without negative-prompt)! Helps in increasing Output Diversity🔥🔥 Negative Token Merging demos: 💡 Self-hosted Gradio Demo 1 (Using FLUX): https://lnkd.in/gsk46gGz ✨ Gradio Demo 2 (Using SDXL): https://lnkd.in/g9RFCXp3 Build "Negative Token Merging" or NegToMe locally with their official Gradio app: https://lnkd.in/gdDG2WDF

🆕 SwiftEdit: Lightning Fast Text-guided Image Editing via 1-step Diffusion 0.23 secs to make edits !!! 🤯 > SwiftEdit is a user-friendly, lightning-fast editing gradio-app that enables instant edits through simple prompts, delivering precise results.👇 Learn more about the project > You can edit facial Identity and expressions via flexible text prompts SwiftEdit for Instant text-guided image editing (in 0.23s)! ✨ Novel contributions: > One-step inversion framework that enables one-step image reconstruction > Mask-guided editing technique to perform localized image editing. SwiftEdit: Lightning Fast Text-guided Image Editing via One-step Diffusion 👉 Useful links: Project: https://lnkd.in/gNVEhW4r Paper: https://lnkd.in/gBQasd-S Code and gradio-app release awaited on Hugging Face, stay tuned! 🤗

🆕 SwiftEdit: Lightning Fast Text-guided Image Editing via 1-step Diffusion 0.23 secs to make edits !!! 🤯 > SwiftEdit is a user-friendly, lightning-fast editing gradio-app that enables instant edits through simple prompts, delivering precise results.👇 Learn more about the project > You can edit facial Identity and expressions via flexible text prompts SwiftEdit for Instant text-guided image editing (in 0.23s)! ✨ Novel contributions: > One-step inversion framework that enables one-step image reconstruction > Mask-guided editing technique to perform localized image editing. SwiftEdit: Lightning Fast Text-guided Image Editing via One-step Diffusion 👉 Useful links: Project: https://lnkd.in/gNVEhW4r Paper: https://lnkd.in/gBQasd-S Code and gradio-app release awaited on Hugging Face, stay tuned! 🤗

Efficient Track Anything Model (EfficientTAM) 📣 🔥 Can run >10 frames per second with reasonable video segmentation performance on iPhone 15. Achieves comparable performance with SAM 2 with improved efficiency (⚡ fast!). Uses a vanilla lightweight ViT image encoder. Try the official 🤗Gradio Demo with a family of EfficientTAMs at: https://lnkd.in/gPT5Jc-M Project page: https://lnkd.in/gTMgN4nw Code: https://lnkd.in/g-sCCwyu Models on Hugging Face Hub: https://lnkd.in/gGgaumXd

📣 Microsoft Research releases Florence-VL, a new family of MLLMs powered by the generative vision foundation model Florence-2. Achieves significant improvements in general VQA, perception, hallucination, OCR, Chart, knowledge-intensive understanding, and more🔥 Useful Hugging Face links to learn more about the model and release:👇 Gradio Multimodal Chat with Florence-VL-8B: https://lnkd.in/gSZbDBbi Florence-VL: Enhancing Vision-Language Models with Generative Vision Encoder and Depth-Breadth Fusion. Paper: https://lnkd.in/grXP_C5b Florence VL 8b SFT Model: https://lnkd.in/gWD5Vhqt

NVComposer is a breakthrough in generating Novel View Synthesis for given image(s) 🔥 > Choose camera movement mode: Spherical Mode & Translation Mode > You can even adjust the spherical parameters and translations along the X, Y, and Z axes > You can upload up to 4 images as input conditions. > Another solid research by Tencent ARC group!🙌 Gradio app for 📸 NVComposer on Hugging Face Spaces: https://lnkd.in/gV5GtZBF

Efficient Track Anything Model (EfficientTAM) 📣 🔥 Can run >10 frames per second with reasonable video segmentation performance on iPhone 15. Achieves comparable performance with SAM 2 with improved efficiency (⚡ fast!). Uses a vanilla lightweight ViT image encoder. Try the official 🤗Gradio Demo with a family of EfficientTAMs at: https://lnkd.in/gPT5Jc-M Project page: https://lnkd.in/gTMgN4nw Code: https://lnkd.in/g-sCCwyu Models on Hugging Face Hub: https://lnkd.in/gGgaumXd

Efficient Track Anything Model (EfficientTAM) 📣 🔥 Can run >10 frames per second with reasonable video segmentation performance on iPhone 15. Achieves comparable performance with SAM 2 with improved efficiency (⚡ fast!). Uses a vanilla lightweight ViT image encoder. Try the official 🤗Gradio Demo with a family of EfficientTAMs at: https://lnkd.in/gPT5Jc-M Project page: https://lnkd.in/gTMgN4nw Code: https://lnkd.in/g-sCCwyu Models on Hugging Face Hub: https://lnkd.in/gGgaumXd

Build AI Agents - simple and fast! 🚀 Using Nvidia NIM, Langchain, and Gradio! Learn more 👇 How to Build a Simple AI Agent in 5 Minutes with NVIDIA NIM: https://lnkd.in/gbiAbmqu

MV-Adapter [Text-to-Multi-View Images] Why generate a single image when you can generate multiple views in one shot? 🔥 🤩 MV-Adapter team has released a Text-conditioned MV-Adapter but specifically for anime style! Check it out 👇 🙌 Kudos to the MV-Adapter team for building and hosting the Gradio app on Hugging Face : https://lnkd.in/gNwe2g6M