Face Recognitions Web App Needed

25% of the total amount will be given to you after you complete 50% of the total work. No amount will be increased for this project at any cost. No project will be awarded or no milestone will be released without checking your work on my project. Our app could be used for eKYC verification using Aadhaar Card Number, PAN Number, Voter Card, Driving License. Where we will upload the image or small video of the i.d card holder into the database and one i.d proof of that person, then app should be able to detect the face of the person and verify that person's photo and his photo i.d proof is same. App should be able to detect colors, weather, activities and moments also through live CCTV Camera feed. Our professional users should be able to use this technology to find suspects within sometime of an incident, out of thousands of hours of footage from diverse camera feeds. This same accuracy is now available to your home and business customers to make their own video footage more useful, usable and relevant. We need these features to be added into our app. We want that app shud be connected to CCTV cameras in Mall and Hotels and it shud recognize the face of visiters and shud tell us the exact count of people visiting there by giving a number to each visitor. There shud be an option of Ekyc verification in the app by connecting some APIs of Driving License of UIDAI. Where an i.d or address proof will be uploaded and it will fetch the information from database and show us the results that this i.d proof and address is genuine or fake. We should be able to have the automatic ability to classify and refine query results by object types, such as people, vehicles, packages, etc. Users should be able to submit queries such as “show me trucks” or “show me people wearing red.” Every new feature creates superb differentiated value, unique to each partner’s needs. Quick review This should not be fast forward: it’s all in one rapid video review. Deep drill down We should be able to Search the day in minutes. Select events and access the original video in a single click. Alert configuration It should Send notification of unusual activity. Gain insights There should be a feature of Graphic visualisation of data analysis (heat maps, pathways, bar graphs, etc.) Share clips Users should be able to One-click export of video clips to email or social media. Handling complex lighting conditions on the fly. Every scene is different- we’ve developed the capacity over the years to handle thousands of scenes. Lighting conditions and weather conditions for example (you’d be surprised how snow and rain impacts the accuracy of computer vision!). Detecting subtle and camouflaged objects. We can identify very small subtle objects, down to the level of small creature. and, while an object in the frame that is appearing against a background of the same color is a difficult task – over the years, our detection has become very sensitive to subtle differences in color and texture. Knowing what goes together to create a single incident. If a person “disappears” from the frame for a period of several seconds, as they pass behind a car or a tree, we detect that and treat it as one incident. If the person goes in and out of the frame boundaries, we treat it as separate object incidents. Extended capabilities through third party metadata. We connect video time stamps to metadata from other systems to create tight custom integrations. For example, if you have an ID badge system in your small business application, we can pair that information to video. The same goes with audio feeds, motion sensors, smoke detectors and more. There should be an admin account and user account. Admin should be able to monitor any user account parallelly. Admin should have rights to add or remove any user. There should be a key from Admin side and after submitting that key only, user account will be activated. We need to embed the app on our drone and will take it towards some public meeting and app shud be able to recognise the number of people available in crowd and recognise their face. Not only do you have the alert from the third party feed, but the video of what happened at that exact moment, along with more efficient presentation and rich query formulation capabilities. The App should be capable to handle following face images- a) Image with minimum face resolution of 50x50 pixels. b) JPEG image format. c) Images with different lighting conditions. d) Images with face angle variation of +/- 30 degrees from frontal in Yaw, Pitch and Roll. e) Face images of people from different race and ethnicity like (but not limited to) North Indian, South Indian, East Indian, Chinese, European, African etc. f) Face images with beards, moustaches, moles, marks and make-up with different ancillaries. g) Face images for different age groups like children, adults and old aged. h) Color images and Black & White images i) Images from different type of cameras for example mobile phone camera, digital cameras, CCTV cameras etc. App should be capable to run with minimum Hardware at Data Centre for at least one Million Image processing i.e. ○ Xeon processor ○ 16 GB RAM ○ 1 TB HDD.