sagieppel

fine-tune-train_segment_anything_2_in_60_lines_of_code. The repository provides code for training/fine tune the Meta Segment Anything Model 2 (SAM 2)

Fully-convolutional-neural-network-FCN-for-semantic-segmentation-Tensorflow-implementation. Python

Fully-convolutional-neural-network-FCN-for-semantic-segmentation-with-pytorch. Fully convolutional neural network (FCN) for pixelwise annotation (semantic segmentation) of images implemented on python pytorch.

Focusing-attention-of-Fully-convolutional-neural-networks-on-Region-of-interest-ROI-input-map-. This project contains code for a fully convolutional neural network (FCN) for semantic segmentation with a region of interest (ROI) map as an additional input (Figure 1). The net receives image and ROI as a binary map with pixels corresponding to ROI marked 1, and produce pixel-wise annotation of the ROI region of the image. The method is based on using region selective features This code was tested on for semantic segmentation task of materials in transparent vessels where the vessel area of the image was set as the ROI.

Train-Semantic-Segmentation-Net-with-Pytorch-In-50-Lines-Of-Code. Train neural network for semantic segmentation (deep lab V3) with pytorch in less then 50 lines of code

Classification-of-object-in-a-specific-image-region-using-a-convolutional-neural-net-with-ROI-mask-a. Classification of object in a specific image region using a convolutional neural net with ROI mask as input

Train_Mask-RCNN-for-object-detection-in_In_60_Lines-of-Code. Train mask RCNN for object detection in pytorch in 60 lines of code

Computer-vision-based-recognition-of-liquid-surface-and-liquid-level-in-transparent-glassware-vessel. Trace liquid surface and liquid level in transparent vessels (python)

Detecting-and-segmenting-and-classifying-materials-inside-vessels-in-images-using-convolutional-net. Detecting and segmenting and classifying materials inside mostly transparent vessels in images using a fully convolutional neural net for chemistry lab and other general setting

Chemical-kinetics-rate-equation-with-tensorflow. Chemical kinetics with tensorflow. Use tensorflow deep learning platform to numerically calculate chemical rate equations for chemical kinetic systems with arbitrary number of reactions and components

Pointer-based-segmentation-find-segment-containing-given-image-point-witin-given-ROI-mask-using-CNN. Category independent sequential image segmentation. Find a segment containing a given image pixel within a given attention mask using a convolutional neural network (CNN), and its application for sequential region-by-region instance-aware segmentation of images with unfamiliar categories.

Generator-evaluator-selector-net-a-modular-approach-for-panoptic-segmentation. Generator Evaluator selector net for image segmentation. Generator net generate Image segments. Evaluator net rank and select the best segments for the image segmentation map.

Unsupervised-extraction-of-textures-and-PBR-materials-from-images. Extract textures and PBR materials from random images using unsupervised statical approach

Segmenting-givne-region-of-an-image-using-neural-net-with-ROI-attention-input. Segmenting specific region of an image using convolutional neural net that receive a container ROI mask and segment the content of this mask in the image. Detecting segmenting liquid materials inside containers for medical and chemistry labs.

Train-neural-net-to-predict-continuous-property-from-an-image-in-40-lines-of-code-with-PyTorch. How to train Resnet neural network to predict continues property (fill level) from image in less than 40 lines of code, using pytorch

Molecular-Dynamic-with-tensorflow. Molecular Dynamic (M.D) with tensorflow: Tensor Graph flow demonstration of the molecular dynamic model

convert-image-into-seamless-tileable-texture. Turn the image of texture, such that it can be tiled seamlessly into uniform texture, and change the boundary area to hide the stitching line in the image native patterns

Simple-template-match-with-rotation. Simple template match with rotation (template crosscorrelation to the edge image of the input image

Using-sobel-edge-detector-on-activation-map-of-convolutional-neural-networks. Use the activation map (or feature map) of convolutional neural net (cnn) as images for sobel edge and gradient detector

Classification-of-the-material-given-region-of-an-image-using-a-convolutional-neural-net-with-attent. Classification of materials given region image region using a convolutional neural net with attention mask as input

Predicting-3D-shape-of-liquid-and-objects-inside-transparent-vessels-as-XYZ-map-from-a-single-image. Given an image containing transparent containers, predict the 3d model and 2D segmentation of the vessel content, the vessel itself, and the vessel opening plane. The prediction is made in the form of an XYZ map that predicts X,Y,Z coordinates per pixel.

Convolutional_Neural_Nets_With_ROI_input_and_region_selective_features_For_Materials_in_vessels. Fully convolutional neural(FCN) that use region of interest (ROI) mask as input and performed semantic segmentation of the ROI region in the image. Using region selective features. Applied for recognition of materials in glass vessels in chemistry lab setting.

Find-object-in-a-image-with-template-match-with-rotation. Find object in a image using template match with rotation

Segment_Anything2_Trained_for_Semantic_Seg_Vessell_Fill_level. Segment Anything Model 2 (SAM2) Trained for Segmenting Fill level/Liquid Level and Vessel

Molecular-Dynamic-With-Pytorch. 2d Molecular dynamic with Pytorch Dnn package include cuda GPU support

Shape-and-Texture-recognition-in-large-vision-language-models-. Coming Soon

Procedural_Annotated_Images_Generation_Liquid_Transperent_Vessel. This script will procedurally generate images of randomly shaped transparent vessel with simulated liquid of random material, and random background # This will be used to generate images that will be saved. This script will generate multiple images in multiple simulation

Geodesic-Distance-on-a-color-image. Find minimal Geodesic distance on an image (Im) between point x0,y0 and all other points in the color image

Zero-Shot-Material-State-Segmentation-Net. Net for Zero-Shot Material segmentation from image. .

MatSim-Dataset-Generator-Scripts-And-Neural-net. Links to MatSim Dataset generator scripts and neural net

Materials-in-Vessels-data-set. Data set of pixel-wise annotated images of materials in glass vessels in chemistry laboratory setting

LabPics-medical-Computer-vision-for-liquid-samples-in-hospitals-and-medical-lab-. code for: Computer vision for liquid samples in hospitals and medical labs using hierarchical image segmentation and relations prediction

Generate_Materials_Recognition_and_Retrieval_Synthetic_Dataset_Blender. Generate synthetic dataset showing the multiple instances of the same PBR material on different objects and enviroments in order to train/test for recognition and retrieval of materials in images, in general settings

Generalized-hough-transform-with-rotation. Generalized hough transform with template rotation

Recognition-of-materials-in-transparent-vessel-using-fully-convolutional-neural-networks. Fully convolutional neural net (FCN) for of recognition and segmentation of materials in transparent vessels, with emphasis on chemistry glassware.

Reconstruct-image-from-sparsely-sampled-pixels-using-fully-convolutional-neural-network-FCN-with-v. Take image in which only small fraction of the pixels are known and reconstruct/upsample the full image using fully convolutional neural nets and valve filters (Tensorflow implementation).

Classification-of-specific-image-region-using-a-convolutional-neural-net-with-ROI-mask-input. Given an image and region of interest (ROI) mask as input, the net classifies the region of the image marked in the input mask.

transform-image-into-seamless-tileable-texture-using-stable-diffusion-inpainting. Change the texture image to seamless, such that if can be tiled seamlessly with itself with no stitching lines. Meaning the opposite side of the image merges seamlessly with each other. This is done by changing the boundary of the image using stable diffusion inpainting option on the boundaries.

MatSeg. MatSeg: Zero-Shot Material State Segmentation Dataset And Trained Net

MatSim-Generator-Generate-image-of-random-materials-on-ranodm-objects-with-Blender. MatSim dataset Generation scripts: Generate image of random materials on ranodm objects with Blender

unet_convnext. Python

Train-pointer-net-for-segmenting-image-into-objects-parts-and-materials-in-60-lines-of-code. Train pointer neural net that segment image for for segmenting image into material phases, objects parts (could also use for panoptic and instance segmentation)

Testing-Large-Vision-Language-Models-LVLM-on-visual-questions. Script for generating visual multiple-choice question by given image with few panels each panel have a letter. In addition, to text. Test script for interface with GPT,LLAma,Gemini,Claude visual quaries.

Geodesic-Distance-Between-Two-Points-On-Image. Find minimal Geodesic distance between point two point (x0,y0 and x1,y1) on image (Im)

TransProteus_Paper_Code. Links To TransProteus Code

Create-a-curvature-factor-map-for-axisymmetric-vessels-based-on-their-outline-in-the-image. Use contour of axissymmetric vessel to create curvature factor map

Reconstruct-image-from-sparsely-sampled-pixels-using-fully-convolutional-neural-network-FCN-. Take image in which only small fraction of the pixels are known and reconstruct (upsample) the full image using fully convolutional neural net (tensorflow implementation).

Find-Geodesic-distance-on-a-image. Find minimal Geodesic distance on an image (Im) between point x0,y0 and all other points in the image (The image can be color or greyscale)

Predict-3D-model-from-image-as-an-XYZ-map. Predict XYZ map from image using neural net

Computer-vision-based-recognition-of-liquid-surface-and-liquid-level-in-transparent-vessel. Computer vision based recognition of liquid level and surface in images

Predict-material-properties-from-image-using-neural-net-in-50-lines-of-code-with-pytorch. Predict material properties from image using neural net in 50 lines of code with Pytorch

Depth_From_Stereo. 3D model from stereo Video

Simple-template-match-with-variable-image-to-template-size-ratio. Find template Itm (binary image) in the canny edge image of image Is (Grayscale image).

Infinite-Landscape-Generator. Generate 3D mesh of landscape/terrain map by extracting height map from images and turning it into ".obj" 3D mesh file that can be used for 3D rendering (bender/unreal engine).