PROJECTS
Mixed Reality Human-Robot Collaboration
Interacting with robots in a collaborative environment often turns out difficult especially when both the humans and the robot desires to work simultaneously on a set of tasks. Hence to overcome this barrier, we are working on a research project based on a mixed-reality communication system, for an intuitive human-robot collaboration. The main objective of this approach is to utilize the physical work environment as a canvas to communicate task-related instructions and robot intentions in the form of visual cues. The system can guide the human while he/she is performing tasks, warn them by clearly displaying the intentions of the robot, thereby creating a more safer working environment. The user can in-turn improve their precision and efficiency.
Team: Yash Rathore, Ramsundar K G, Heni Ben Amor
#C++ #OpenSceneGaph #OpenCV #Eclipse #VisualStudio
Blind Dating: Making Mobile Dating Accessible for People with Visual Impairments
Online dating platforms such as Tinder, Bumble, and Coffee Meets Bagel, have become culturally accepted methods of meeting new people. However, many of these online dating platform’s user profiles rely heavily on visual data. For individuals with visual impairments, the way that these applications are designed presents a barrier to the opportunity for them to form new relationships. Using voice-over with any of these apps creates profound confusion as they are not built with accessibility in mind. We proposed a method of providing users with visual impairments access to the data encoded in the images of other user’s profiles.
To find what information people care about knowing the most in a dating profile, we surveyed the public and then built simple classifiers for the most popular features that were identified by the survey. As the APIs for the dating apps mentioned above are not public, we created an application that works via screenshots. When the user would like to know more about a given profile, they just need to activate the native screenshot functionality, and our application will find the screenshot in the screenshot folder on the phone, send it to our web service where the image is analyzed and output text is formed. This text is sent back to the phone and read out, providing the user with a coarse idea of what the image contained.
Team: Corey Heath, Meredith Moore, Yash Rathore, Anusha Rajan, Guru V, Pawan O
#AndroidStudio #Linux #Python #OpenCV #Cloud9
Astra: A VR+LeapMotion shooting game
Astra a.k.a अस्त्र (a supernatural weapon) is an immersive VR game which enables users to use gestures to shoot at asteroids and score points. Astra is a combination of VR and Leap Motion. VR is used for head tracking and visual immersion whereas Leap Motion is used to detect gestures.
Team: Nayan Seth, Yash Rathore, Aakash Chaurasia, Akriti Jain, Anusha Rajan.
GitHub: link
#Unity3d #HTCVive #LeapMotion #Csharp
Treasure Raider: Android game to unlock treasure mapped onto real world
Mobile games these days are fun to play and some of them even end up in a short term addiction for the users. When these games are mapped onto real world,it takes mobile gaming to a whole new level. Games like PokemonGo,Ingress,Parallel Kingdom, etc. were one of the most successful games in this domain. Inspired by these ideas we developed a location based realworld Treasure hunting game where players solve mysteries and physically visit various locations in order to complete the game. GPStracking of the players and realtime navigation system are one of the key features involved.
Team:Aakash Chaurasia, Yash Rathore, Akriti Jain, Anusha Rajan.
#AndroidStudio #Java #GoogleMaps #GeoLocation #GPSNavigation #FirebaseAuth
Gesture Interactive Board: Making traditional classroom teaching more fun
With the growing advances in information technology and computer visualization, we have come to a stage where it is feasible to completely virtualize the classroom teaching methodology using sophisticated displays having multi-touch gesture recognizing capabilities, but the question is ,Can common people afford such technology? Clearly, the answer is no, in a country like India with current economic, population and the number of educational institutions. Our aim is to come up with an affordable substitute for such technology, ‘A digital Interactive Board’ which will recognize hand gesture and give an intuitive means of teaching to students.
Team: Yash Rathore, Fayaz Ali.
Disease Detection and Classification in Soya Plants.
Monitoring crops for diseases detection require a lot of human expertise. However this might be prohibitively expensive in large farms and even may not be available in remote places. To overcome this issue and increase the productivity of Soya bean farms we developed a computer vision based mechanism to monitor plants for leaf diseases. The images of infected leaves are processed to obtain textural and color features and are fed to a classifier to detect the disease type. The machine classifier is trained with large number of labeled training samples.
- The main objective of this project was to detect and then classify the various diseases in Soya bean plant using Computer Vision and Soft Computing techniques where the input is in the form of visual data.
- Secondly, we tried using different methods for classification and choose the most efficient one based on the compared results.
#Matlab #OpenCV
