Ömer Karışman

omni-zero. A diffusers pipeline for zero shot stylised portrait creation

omni-zero-couples. A diffusers pipeline for zero shot stylised couples portrait creation

grounded-segmentation. A streamlined implementation of Grounding DINO and SAM for advanced image segmentation. This lightweight solution simplifies the integration of powerful object detection and segmentation models, offering an easy-to-use interface for developers seeking efficient image analysis capabilities without complex setups.

simple-lama. Simple LaMa Inpainting: An easy-to-use implementation of the LaMa (Large Mask) inpainting model. Remove unwanted objects or fill in missing areas in images with just a few lines of code. Supports both CPU and GPU processing, and integrates seamlessly with HuggingFace Hub for model loading.

head-segmentation. A lightweight implementation of head segmentation using a pre-trained model, ideal for quick data preparation and image processing tasks.

roop. one-click deepfake (face swap)

Times. [Swift] Control how many times a codeblock gets executed.

p2p-diffusion.github.io. Project Site for Person2Person Diffusion

airchat. Objective-C

replicate-dreambooth-model-pusher. Feeds the Replicate DreamBooth API back with the input weights so it can do its magic and push a trained model for you.

PhotoMaker. PhotoMaker

VibeVoiceCommunity. Long-form conversational TTS | Community fork

ReliableSwap. Official Implementation of 'ReliableSwap: Boosting General Face Swapping Via Reliable Supervision'

StyleFace. A web application to generate and alter faces using Generative adversarial networks.

kling. Python

controlnet_aux. Python

StableTwoUnet. Jupyter Notebook

face-alignment. :fire: 2D and 3D Face alignment library build using pytorch

layer-diffusion. Unofficial implementation of Layer Diffuse in diffusers

GFPGAN. GFPGAN aims at developing Practical Algorithms for Real-world Face Restoration.

BasicSR. Open Source Image and Video Restoration Toolbox for Super-resolution, Denoise, Deblurring, etc. Currently, it includes EDSR, RCAN, SRResNet, SRGAN, ESRGAN, EDVR, BasicVSR, SwinIR, ECBSR, etc. Also support StyleGAN2, DFDNet.

VibeVoice. Frontier Open-Source Text-to-Speech

grounded_sam_replicate. Implementation of Grounding DINO & Segment Anything, and it allows masking based on prompt, which is useful for programmed inpainting.

cocomit. Estimate the cost of each Git commit using the COCOMO model.

photalabs.

jsont. Go