Event ended

Wolves Summit AI Hackathon Summary

Wolves Summit AI Hackathon image

Hackathon Overview

Our AI hackathon brought together a diverse group of participants, who collaborated to develop a variety of impressive projects based on:

916

Participants

24

Teams

6

AI Applications

Winners and Finalists

medal

Chase The Fire

Our solution consists of 2 parts, before the forest fire and after the forest fire. Before the forest fire First protocol (software solution): We used M.L. To solve the problem in particular library scikit-learn inside Python and with a dataset containing scripts we used NLP to convert the string to an integer so that the algorithm could handle the dataset and then split the data into partial training and partial test. They were divided into five classification algorithms and we compared the results on their own by examining the model to find the best algorithm for dealing with this data and then using the best algorithm in the metrics model. Then we saved the model and its weights. We have built an application with pyqt5 which receives potentially fire factors from datasets and sensors to predict the fire before it happens. After the forest fire Second protocol (hardware solution): The console makes the user feel extra confident. It is connected to a solar panel that automatically supplies power to the system. The controller will be positioned all over the forest which sends alarms and data to the server which throws the GSM system e.g., temperature, humidity, smoke, fire detection, fire location, rain detection, and wind direction for fire tracking. As well as making a plan for a safe path for animals in case of emergency by buzzing or whistling to keep the animals away from danger. This data is taken to create a new dataset through which the system is updated ml algorithms and the past dataset.

Chasers

medal

fire detector

The app designed by team innovators is a wildfire detector and risk analysis. This project has two branches in which the first one is wildfire detector which is build using deep learning model. The detector asks for a forest image and predicts if there is fire or no fire in the image. The second branch is of risk analysis which is build using a machine learning model. In here we are predicting the next day wildfire spread. This app can be a great asset for researchers and climate change organizations to be prepared to fight the wild fires. Most of the description is provided in the presentation and the github repository. For demo of the project look into the repository on github we have added the demo links in the read.me. That's all! Thankyou!

Innovators

medal

SOILTREE

The past decade has been onerous to global humanity. The COVID-19 pandemic and triple global crisis undermine decades of global progress. Nevertheless, these events drew global attention to the Sustainable Development Goals. Deforestation is definitely one of the world's major concern. However, we as human beings require wood for constructions and manufactures. Annually, deforestation timber also generated approximately $500 million in economic value. Since we can't stop the trees from being cut down, we can plant fast growing timbers on suitable soils. How should we know which type of timbers should we plant based on the soil? Here we introduce a timber guide system to recommend the best timber to be planted.

Crops Analysis

This event has now ended, but you can still register for upcoming events on lablab.ai. We look forward to seeing you at the next one!

Checkout Upcoming Events →

Submitted Concepts, Prototypes and Pitches

Submissions from the teams participating in the Wolves Summit AI Hackathon event and making it to the end 👊

Help to spread the word and share these amazing projects!

Fake smile detection

Fake smile is an emotional sign on the face that can be used as information for non-verbal communication. One of its functions is for lie detection purpose based on the information of emotional sign generated on the face. The emergence of fake smile indicates that there are negative emotions, uncomfortable feeling, and something hidden in a person. This research aims to detect fake smile. In fact, real smile is characterized by the contraction of zygomatic major muscle on the edge of mouth corner and obicularis oculli muscle on the eyelids. However, on a fake smile, zygomatic major muscle experiences contraction, but obicularis oculli muscle doesn't contract. Contraction of the zygomatic major muscle is identified by the appearance of wrinkles on the cheeks corner of the mouth, whereas obicularis oculli contraction is identified by the feature value of eye elongation. On the test image, segmentation of RoI (Region of Interest) is done on cheeks and eyes. With the growing era of social media, it is difficult to identify the real from fake whether it is any news or face/video of any celebrity, politician, etc. Also, the fake or manipulated faces and videos are being generated enormously which are harder to detect by traditional means of software or methods.

Trojan Horses

medal

Chase The Fire

Our solution consists of 2 parts, before the forest fire and after the forest fire. Before the forest fire First protocol (software solution): We used M.L. To solve the problem in particular library scikit-learn inside Python and with a dataset containing scripts we used NLP to convert the string to an integer so that the algorithm could handle the dataset and then split the data into partial training and partial test. They were divided into five classification algorithms and we compared the results on their own by examining the model to find the best algorithm for dealing with this data and then using the best algorithm in the metrics model. Then we saved the model and its weights. We have built an application with pyqt5 which receives potentially fire factors from datasets and sensors to predict the fire before it happens. After the forest fire Second protocol (hardware solution): The console makes the user feel extra confident. It is connected to a solar panel that automatically supplies power to the system. The controller will be positioned all over the forest which sends alarms and data to the server which throws the GSM system e.g., temperature, humidity, smoke, fire detection, fire location, rain detection, and wind direction for fire tracking. As well as making a plan for a safe path for animals in case of emergency by buzzing or whistling to keep the animals away from danger. This data is taken to create a new dataset through which the system is updated ml algorithms and the past dataset.

Chasers

TrashCollector

During recent years, people started to produce more and more trash, polluting the planet. Even with continuing recycling, there is one important issue – trash on streets. It can be in already packaged bags or individual wrappers and bottles on the group, with some of it later being in the waters of Earth. The idea of the project is to create a trash detector using computer vision networks that can be used in 2 ways. Firstly, it can be integrated with already installed security cameras and have some software for trash collectors to use, so they would know where the trash is. Second more improved strategy is to create an automated robot which would collect the trash.

EarthAI

medal

SOILTREE

The past decade has been onerous to global humanity. The COVID-19 pandemic and triple global crisis undermine decades of global progress. Nevertheless, these events drew global attention to the Sustainable Development Goals. Deforestation is definitely one of the world's major concern. However, we as human beings require wood for constructions and manufactures. Annually, deforestation timber also generated approximately $500 million in economic value. Since we can't stop the trees from being cut down, we can plant fast growing timbers on suitable soils. How should we know which type of timbers should we plant based on the soil? Here we introduce a timber guide system to recommend the best timber to be planted.

Crops Analysis

medal

fire detector

The app designed by team innovators is a wildfire detector and risk analysis. This project has two branches in which the first one is wildfire detector which is build using deep learning model. The detector asks for a forest image and predicts if there is fire or no fire in the image. The second branch is of risk analysis which is build using a machine learning model. In here we are predicting the next day wildfire spread. This app can be a great asset for researchers and climate change organizations to be prepared to fight the wild fires. Most of the description is provided in the presentation and the github repository. For demo of the project look into the repository on github we have added the demo links in the read.me. That's all! Thankyou!

Innovators

Sustainable Journey Assistant

The automotive industry is one of the largest industries in the world by the revenue it generates. As technology advances the opportunities, advancement, and challenges increase multifold in the industry across the globe. .Traffic jams and congestion each year costs the world economy over a trillion dollar Engine Idling in jams contribute to harmful particle emissions that also get concentrated more near the location . Besides, The vehicles of today and the future require highly versatile onboard diagnostic systems that can detect and recognize errors and possible malfunctions. There is a huge demand for new and improved solutions that can help achieve this. Our goal is to create an IoT solution, which displays different methods of transportation, calculates their carbon emissions, and recommends the most eco-friendly routes – all in one application. With a touch of a button, commuters will get the chance to make smarter and more sustainable decisions when it comes to their travel journey. We aim to connect different modes of transportation in one app as a connected ecosystem and calculate carbon footprint based on the transportation journey. Using this we educate consumers on what their carbon footprint means and empower them to change towards more sustainable habits and enable payments towards different modes of transportation to happen seamlessly in one app. For the diagnostics section 3D Model of the car will have multiple points on it which will get highlighted based on the fault so as for the user to understand it. Historical data of users and the fault they faced will be stored using their consent which will help the future users .Helpers would be incentivized using reward system and with growing users the proximity of a user helping a stuck car would increase. While the car enters an area with low connectivity , it would prompt the user to download the essential data for offline use. We offer On Board Diagnostics Scanner to our user for FREE so that errors can be found more frequently. The idea is convert empathy into actions.Our app would arrive at various results and then convert it into actions.The first one being the speed optimizer.Traffic jams and congestion each year costs over a trillion dollar.Engine Idling in jams contribute harmful particle emissions.The app which would display an "IDEAL SPEED" for the vehicle(from varios factors) ,driving at which the driver won’t face any traffic jam thus preventing engine idling, less pollution and preventing mental stress.The app would take data like car’s speed,nearest jam/congestion using location data etc,keep updating itself on user actions.Further,our common payment plan linking public and private transportation on pay-later cycle gives user insights where he could save money,carbonfootprints and empathise on issue. The app arrives at meaningful facts to which user can act on to.It connects the user on points like lessening car congestions , car pooling etc.The cost of wasted fuel can add up to $70-650 a year, depending on fuel prices, idling habits and vehicle type).Idling actually increases overall engine wear by causing the car to operate for longer than necessary.For every 10 minutes our engine is off, we prevent one pound of carbon dioxide from being released (carbon dioxide is the primary contributor to global warming) 11 million tons of carbon dioxide, 55,000 tons of nitrogen oxides, and 400 tons of particulate matter are emitted into the environment from heavy-duty truck idling during rest periods.Emissions from engine idling are contributing to climate change and impacting air quality, which pose negative health risks for everyone. We introduce a “Pay Later” feature in the app where the user can sync all his transportation modes and we suggest methods to lessen the monthly cost based on the use pattern and where the person could have improved by adhering to the suggestions like car pooling , public transports etc. The user does the payment on a 15 Day cycle and we provide insights to the user based on his use and suggest the methods of lessening the carbon footprint. Further, if the user obeys the tips and suggestion , we reward him/her with third party offers/coupons .Revenue generation can be made at this node by partnering with brands on this.Further, our common payment plan linking public and private transportation on pay-later cycle gives user insights where he could save money , carbon footprints and empathize on issue. We can provide the Convenience of 1-Tap Check Out and give analysis to the users on their usage .

BharatVerse Innovations