nehay06 Goto Github PK

angrybirds

computer-vision-blobdetection

To detect blobs at multiple scales, there are two strategies to complete the task: (1) iteratively filter the image with kernel of increasing size. (2) iteratively downsample the image and filter it with the kernel of fixed size.

computer-vision-demosaicingimage

in digital cameras the red, blue, and green sensors are interlaced in the Bayer pattern (Figure 2). The missing values are interpolated to obtain a full color image

computer-vision-image-whitebalance

White Balance is based on \gray world" assumption that states the average color of the image under white light is gray

computer-vision-imagealignment-prokudin-gorskii-collections

IMAGE ALIGNMENT The images are taken vertically on glass plates with the order blue, green and red from the top. To align the image, first it is cut into three sections, each representing a color channel. Then the blue and red channels are aligned to the green channel.

computer-vision-imagefiltering-hybridimages-master

Computer-Vision-ImageFiltering-HybridImages A hybrid image is a combination of a low-pass filtered (i.e. blurry) image and a high-pass filtered (i.e. sharp) image. Recall that one can obtain a sharp image by subtracting the blurry version of an image from itself.

computer-vision-panamoric-stiching

computer-vision-shape-from-shading-photometric-stereo

Photometric stereo The input to the algorithm is a set of photographs taken in known lighting directions and the output of the algorithm is the albedo (paint), normal direction, and the height map.

crossdomainimagematching

Images of same objects across different visual domains like sketches, paintings, photographs in varied environment, tend to show large difference at raw pixel level.This makes it difficult to match such images. We propose a simple and efficient method to compute cross domain image similarity by utilizing standard object detection framework to find relative importance of features. These weighted features are then used to find image matches, based on visual and contextual similarity. For robust experimentation analysis, we test our approach on UFU-DDD[6], PASCALVOC[ 18], INRIA Holidays[19], Random Flickr images[20] and Sketchy datasets[21].

deeptransitiondepparser

Dependency parser

indexer

Create a Retrieval API that enables performing query retrieval on your index

internet-of-things--part-1

internet-of-things-part-2

moviegenre

paint_project

quoraquestionpairs

Predicting whether two sentences are a paraphrase of one another is an active re- search problem with many real life ap- plications. One such application is min- imizing duplicate questions and answers on question answering websites such as Quora or Yahoo! Answers. We used the Quora Question pairs dataset (Iyer et al., 2017) which consists of question pairs tagged as duplicate and not dupli- cate. Among the methods tried, the best accuracy achieved was 82%.