You've landed at the right place. oDesk is now Upwork. Learn about the new platform.

Computer Vision Jobs

11 were found based on your criteria

show all
show all
only
only
only
show all
only
only
only
only
only
show all
only
only
only

Hourly - Est. Time: Less than 1 week, Less than 10 hrs/week - Posted
I'm looking to have an exact replica of http://people.csail.mit.edu/cchoi/research_rgbdtracking.html Utilizing the assets they used (provided on that page) Not in real time, and with no focus on speed performance or efficiency. Just functionality, within at least 90% of the same accuracy that the paper describes. People who have worked with machine vision and algorithmics only please. Spam applications will be reported and marked as spam.

Hourly - Est. Time: 3 to 6 months, Less than 10 hrs/week - Posted
We need a machine vision developer - specifically an OpenCV guru with some background in deep-learning toolkits - on an ongoing basis for various projects. - Open CV and/or other machine vision toolkits - GPU programming CUDA and/or OpenCL - C++ wizard , (Java or Python nice-to-have) - Video understanding, semantic segmentation, Action recognition, SLAM, and efficient flow/stereo methods - ANNs, DBN, CNNs, RBMs, “deep-learning” with frameworks such as Caffe highly desirable - Knowledge of machine-vision approaches highly desirable Experience designing and implementing real-world, high-performance machine learning systems. With your application please submit a brief summary of the hardest problem you worked in your career and why you consider it your crowning glory achievement. Note: This post is NOT for the amateur who has done a term project with machine vision, we are looking for the most skilled resources, **Spam applications will be deleted immediately. Don't even...

Fixed-Price - Est. Budget: $ 500 Posted
Develop software using python and openCV to create a depth map of a person’s face and torso. You can MATLAB code we already developed as guide. Experience required - Python - image processing - openCV - face detection Preferred: - Experience with MATLAB to look at existing code Details: https://docs.google.com/document/d/1v2c4kpVr-A_bp7E9k4zNXvofp2y-uuhu3_eW_znQVfw/edit?usp=sharing Budget: At least $500, please provide your estimate and time-frame Can lead to long-term position if goes well

Fixed-Price - Est. Budget: $ 400 Posted
We have developed a method for identifying polygon objects and recoloring their surface while preserving lighting conditions. This is a two part algorithm. Part one requires no user input, runs a custom line detection method, and finds all line intersections. Part 2a has a single coordinate as user input, extracts a color from that coordinate and finds a polygon around that coordinate using both the custom line detection from part one and color segmentation. 2b takes a user input color, maps the image to a different colorspace and changes the polygon around the user input to the target color to preserve shaging of the original image. Please note that the lightness of the input color is considered an important part of that color and cannot be separated from the input. We have a process in this algorithm which sometimes does not appear end and thus causes a drain on memory (perhaps a memory leak, not certain). As our application scales, this causes the server to run out of...

Hourly - Est. Time: 1 to 3 months, Less than 10 hrs/week - Posted
The scope of this project is to develop an image processing system for a pipe welding robot. The robot control system will rely on feedback from the image processing algorithms to determine the location of the welding electrode in relation to the weld seam edges to correct positional errors. It is anticipated that to achieve these objectives the following general steps will be required: 1) Establish a coordinate system 2) Calibrate the image 3) Window the image (for weld pool and weld seam edges) 4) Remove interlacing artefacts 5) Apply a median filter 6) Apply a sharpening filter 7) Identify edges using a modified Canny operator 8) Remove false edges 9) Scan for edges 10) Fit functions to the edges 11) Determine the positional error 12) Interface to the control system Refer attached illustrations for further information.

Hourly - Est. Time: 1 to 3 months, 10-30 hrs/week - Posted
In order to ensure you have read through this job listing, please include the word “yellow” as the first word in your reply. OVERVIEW: Hey there. I’m a programmer and computer vision researcher who is putting together a set of deep learning and convolutional neural network tutorials to help other coders more easily learn and apply deep learning. I am covering much of the “code example” tutorials, but I really need some help writing out the more technical content and explanations. I am looking to have 5-7 tutorials written covering the following topics: 1. An introduction to deep learning 2. A tutorial on Deep Belief Networks 3. A tutorial on Convolutional Neural Networks 4. A tutorial on what each layer of a CNN is used for, how to utilize each layer, and when each layer should be utilized. 5. A tutorial demonstrating how to train a model using Caffe. 6. A sort of “What’s next?” tutorial where you can point out resources or recommendations moving forward. ARTICLE LENGTH...

Fixed-Price - Est. Budget: $ 3,000 Posted
We have a project to count vehicles using CCTV video feeds in an outdoor setting. This is how the project is envisioned: (1) Install CCTV camera and send to a mini PC server housed together on the advertising panel (2) The video feeds will be analyzed in a dedicated software (can refer to any OpenCV project to count vehicles) (3) The count analytics will be sent to cloud via 3G connection OpenCV Reference: https://www.behance.net/gallery/Vehicle-Detection-Tracking-and-Counting/4057777

Hourly - Est. Time: Less than 1 month, Less than 10 hrs/week - Posted
We are looking for a Computer Vision expert to work with our team to advance ongoing work registering multi-modal sensors to a common real-world coordinate system. This project will leverage registration work already done to co-locate three sensors and bring in a fourth. The application is physical security. We anticipate the initial time commitment to be approximately 80 hours at a full or half-time rate. We plan to be flexible and this project could extend into additional work.

Fixed-Price - Est. Budget: $ 120 Posted
I'm trying to create a C# program, which will build, train and fit Active Shape and Active Appearance Models. To do this, I read a few chapters of T.F. Cootes's and C.J. Taylor's article "Statistical Models of Appearance for Computer Vision". From the beginning it was relatively simple and straightforward, but then began to appear some formulas which were very abstract, became not understandable what data is necessary to substitute in formulas and in which form, began to appear inaccuracies(probably, inaccurate to me, as a layman in this kind of data analysis), as well as some of the details began to overlook, because of which meaning became more abstract, ambiguous and confusing. Due to this it became necessary to fully understand the algorithm of building, training and fitting Active Shape and Active Appearance Models. I believe that the right thing is to be done in UML form or other type of algorithm description (this can be your proposal). The main thing is not the form and presentation...