Computer Vision Jobs

15 were found based on your criteria {{ paging.total|number:0 }} were found based on your criteria

show all
  • Hourly ({{ jobTypeController.getFacetCount("0")|number:0}})
  • Fixed Price ({{ jobTypeController.getFacetCount("1")|number:0}})
Fixed-Price - Intermediate ($$) - Est. Budget: $1,000 - Posted
I need an API created that will take a metadata definition and input photos and create a single photo output that looks as natural as possible. See the attached images for the rough idea. Each class of recipe (e.g., Pizza) will have its own structure, such as: Dough Base - white, wheat, green Sauce - marinara, pesto, white Cheese - mozzarella Toppings[up to 3] - onion, mushroom, etc. Then it would be given the ordering and images (many images would have quite a bit of transparency) and whatever other reasonable metadata that could be provided, and then create the output. There will be thousands of different types of food (not just pizza!). I have done experiments with straight compositing. It doesn't look horrible at low res, but the lack of shadows made everything look artificial. Since this is food, it should look good and as photorealistic as possible. In other words, it has to be smarter than simple compositing. I don't quite know how to write the requirements for this project, so I'm open to clarifying questions.
Skills: Computer vision Image Editing Image Processing
Fixed-Price - Expert ($$$) - Est. Budget: $500 - Posted
We have secured a government contract for automated traffic monitoring system. We want to monitor automated traffic violation detection as accurately as possible. Our first milestone is very simple and should consist of the following rules: 1. Detect passing on red light 2. Detect line crossing. 3. Detect vehicle opposite direction movement 4. Detect vehicle-vehicle and vehicle-pedestrian collision 5. Detect when vehicle doesn't let pedestrian pass. We need to achieve this milestone as soon as possible, after that we will have additional triggers to work on, and we expect the candidate winning this bid to help us further. We don't ask for any specific technology/language to use, it is up to the candidate to decide. If you have enough experience working with computer vision in python you can work with python. Salary and deadline is negotiable. We expect that you tell us how much the project will cost in terms of time and money. PLEASE NOTE that only candidates who will be able to demonstrate prior experience working with computer vision, object detection and/or machine learning will be considered. Please write your ideas/visions how you are going to achieve the goal, which technologies you will use, what can be the main risks and what kind of error rates to expect in violation detection. UPDATE I made a bad job describing my intent. the compensation (500$) is not a fixed price, I'm willing to pay how much you ask (withing reasonable limit of course) as long as you deliver working solution. I didn't do enough research to evaluate the cost, so I expect the candidate will know it's job enough to do the actual estimation.
Skills: Computer vision Machine learning OpenCV
Fixed-Price - Entry Level ($) - Est. Budget: $120 - Posted
Hi you need to implement the pipeline in this paper to get results of classification, but most of the parts of the codes are available and just needs to connected together. http://www.cv-foundation.org/openaccess/content_cvpr_2015/papers/Liu_Towards_3D_Object_2015_CVPR_paper.pdf You need to implement algorithm in section 4, if you look at figure 1, at the deep bimodal boltzmann machine section, but i think mostly the code is available. It takes data from the previous step and the outcome of the deep bimodal boltzmann is fed into the Examplar SVM. The code for the first part is available here https://github.com/s-gupta/rcnn-depth the last step is available here https://github.com/quantombone/exemplarsvm the middle part you need to implement deep bimodal distribution in section 4. There is already code similar to this that will help alot, from here http://www.cs.toronto.edu/~rsalakhu/DBM.html https://github.com/nitishsrivastava/deepnet
Skills: Computer vision Machine learning MATLAB
Fixed-Price - Intermediate ($$) - Est. Budget: $400 - Posted
Looking for a .NET expert who can go ahead and dig into the basics of computer vision library "Emgu CV". You will be required to study the relevant functions for face detection and face identification. This needs to be a Windows form application with MySQL as database. Following steps will be needed: 1) Face detection from webcam, live streaming, always on (already done) 2) User registration followed by Training the application for his/her particular face and generate xml from training set. Generic XML may also be used if success rates are optimal. 3) Face recognition (Display a rectangle on face with his/her name and a success message. Also create a log in the database with time stamp and name of the user), and call API of our backend to mark confirmation. P.S.: This can be developed in native OpenCV as well, but both desktop and web based applications should be able to interact with it. Just for information, desktop application handles a USB webcam for live video stream capture (which needs to be developed as a part of the project) and web application is concerned with data analytics and reporting (not a part of the project). I am open to all feasible solutions. Please share your thoughts over the same with approach strategy that you will be adopting. Thanks Zeeshan ---- UPDATES: There is no need to handle database directly. I have an existing we based application which has API for login, mark attendance, get attendance sheet etc. which you can use in this software. Please go through attached file for exact details. You can update your bid accordingly if required.
Skills: Computer vision .NET Framework OpenCV
Fixed-Price - Entry Level ($) - Est. Budget: $500 - Posted
We need a developer who has experience using Google Cloud Vision API for labeling of images. The python script will need to do the following: 1. Access a large collection of images via a REST API. 2. Extract a list of keywords per image from Google Cloud Vision API. 2.1. We will use Google's trained classifier. 2.2. We will need the performance to be reasonable as it is meant to be used on data sets of millions of images. 2.3 This is not a must but an advantage if you have experience using a custom classifier. 3. Post back the list of keywords via REST API. The acceptance criteria will be running correctly on our test database of 20,000 images. The script should be as simple as possible. Experience with computer vision is not a must but is a big plus. More details will be provided to the right applicants.
Skills: Computer vision Python
Fixed-Price - Expert ($$$) - Est. Budget: $250 - Posted
I'm working on this paper, multispectral stereo odometry, i need a reviews of code, because there is something that is wrong. I'm not able to get the correct trajectory. I would want that you correct my code ,and a matlab script to plot trajectory. paper: https://mega.nz/#!eMVSiaiD!qzaRTuU_jmx_gmTCROqrybmsJePEreLIjFvfac5gD5E second paper on which, the one is based ,it is important just for motion estimation, because they use the same motion estimation method: https://mega.nz/#!rZUjlRKa!EZzSpmiu_bcs00SUTTZwnVo6qKBk9uxUyBPbqnvQhw4 Please before to ask job , read paper ,if you are able to solve it, I will send code and dataset and further information ,you will need
Skills: Computer vision C++ Image Processing OpenCV
Fixed-Price - Intermediate ($$) - Est. Budget: $1,000 - Posted
I need a utility that will process an image and crop to a portion of the image. The domain is collectible sports cards where I need to crop out extraneous content that is not the actual card itself. Additionally, I need some rough measure of quality so that for a given set of images of the same card I can identify the "best" one (e.g., least skew). I will send more detailed requirements, but I've attached an image to provide a few examples of my goal as a starting point so you can evaluate the feasibility of the project. There are approximately 100,000 of these cards. Notes for image: * All card images will be of cards enclosed in a plastic holder with a title rectangle (the format/color of this title area varies, but is always a rectangle at the top). * Some cards will be slightly rotated and/or skewed * The card designs themselves vary quite a bit. Some have a high-contrast background; others have a rule around the edge; some have a more subtle difference with the non-card background.
Skills: Computer vision Image Processing
Fixed-Price - Intermediate ($$) - Est. Budget: $250 - Posted
I am looking to create a facial recognition script to be run in a virtual machine using C++. I would like for the facial recognition script to take the image(s) of a scene and create a unique ID for each face recognized. That unique ID should be able to be recreated by another picture with the same face. In other words, I should be able to take that ID and match it to that person in any image they are in. Preferably using OpenCV unless you have a better suggestion.
Skills: Computer vision Machine learning OpenCV
Fixed-Price - Expert ($$$) - Est. Budget: $200 - Posted
1) Need to extract 7 fields from pdf invoices from USA and Canada. Fields are Invoice\Bill#, Vendor Name, Date, Due Date, Sub Total, Tax, Total and store them in an excel sheet. 2) Searchable text based PDF documents accuracy has to be 100%. 3) It will be a service that will integrate with a web application (MVC with asp.net).
Skills: Computer vision C# Image Processing