First Prize

AK PAL ROBOTIC
PROJECT - ARTIFICIAL INTELLEGENCE & MACHINE LEARNING. BUILT A AUTOMATIC ROBOTIC ARM LOCOMOTIVE PLATFORM.
 

Second Prize

TECH9
PROJECT - SMART STICK FOR DISABLED - BUILT A WALKING STICK FOR SPECIALLY ABLED PEOPLE USING AI, ROBOTICS, ADVANCED MACHINE LEARNING, SENSORS & IOT

Third Prize

SWIFT ACE
PROJECT - BHAMASHAH LOGIN USING SINGLE SIGN ON FACIAL AUTHENTICATION. USED BHAMASHAH API
 

S No. Team Name Category City Theme Idea Prizes
1
BRIDGEi2i Solvers Kumar Vivek
Arpan Naik
Sreekiran
Shashank Rusia
Best in Blockchain Technology Bangalore/Bengaluru Internet of Things

Water Monitoring and Distribution system: Design and develop a low cost reliable and efficient technique to make proper water distribution by continuous monitoring and also controlling it from a central server so that we can solve water related problems. Electricity saving using Light controlled Street Light system: Making streets brighter at night, improving security for pedestrians and drivers, increasing the feeling of safety for citizens, and beautifying the city with a nice and warm atmosphere until dawn.
...Read More

ipad
2
NULL Pointers Ujwal P
Akash James
Sai Somanath Komanduri
Mohit Kumar
Best in Hardware Bengaluru Machine Learning

Build a smart wearable device that uses Machine Learning models to aid the visually challenged. We aim to build a device that uses Concurrent Neural Networks (CNNs) to detect objects and help the visually impaired by notifying them of their surroundings through audio.
...Read More

ipad
3
Digisthan Yashwant Bokadia
Jugal Porwal
Pranay Mann
Dhruv Kumar Joshi
Best In Community Impact Udaipur Artificial Intelligence

We plan to make RPA (Robotic process automation) that automate the claim (Eg. Pension Schemes and Scholarship) verification process which will be regulated by latest government policies and it will ensure fast transparent approvals to the citizens, hence reduce corruption, delays, human bias and error.
...Read More

ipad
4
TecRush Ayush Agrwal
Sohil Shaikh
Akash Hadke
Prafful Awaghade
Best in Internet of Things Pune Internet of Things

TOPIC :- SMART GAS ALERT SYSTEM (SGAS). Patent Application number : 201721031951 [TEMP/E-1/32642/2017-MUM] SOLUTION :- This system helps to detect and send notification of gas leakage, fire event, statistics of weekly and monthly usage of gas, gas weight level detection to users. Users can easily access data using android app, web and access through cloud server. All information about usage of gas, gas booking, and current gas weight will automatically store on central cloud server. We have to set threshold value for gas weight. Once it cross the threshold value of gas usages, then its set maximum gas utilization bit. SGAS send alert as a SMS or GPRS packet to the users for next gas booking or it provides auto gas booking system to the users. So SGAS is a SMART technique to reduce human death ratio due to the gas leakage. KEYWORDS: LPG, Gas Leakage, Fire Detection, Maximum gas utilization bit, Cloud.
...Read More

ipad
5
PFET Animesh Gupta
Raghav Chawla
Amit Manchanda
Aman Shivhare
Best in Machine Learning Haridwar Machine Learning

We can divide our problem into three sub-problems. First, a computer vision problem of understanding the given scene (i.e. in this case, the GUI image) and inferring the objects present, their identities, positions, and poses (i.e. buttons, labels, element containers). Second, a language modeling problem of understanding text (i.e. in this case, computer code) and generating syntactically and semantically correct samples. Finally, the last challenge is to use the solutions to both previous sub-problems by exploiting the latent variables inferred from scene understanding to generate corresponding textual descriptions (i.e. computer code rather than English) of the objects epresented by these variables. Our approach is based on Convolutional and Recurrent Neural Networks allowing the generation of computer tokens from a single GUI screenshot as input. These approaches have the advantage of being differentiable end-to-end, thus allowing the use of gradient descent for optimization. During training, the GUI image is encoded by a CNN-based vision model; the context (i.e. a sequence of one-hot encoded tokens corresponding to DSL code) is encoded by a language model consisting of a stack of LSTM layers.
...Read More

ipad