leadingindiaai Goto Github PK

using-dcgan-to-generate-faces-of-animated-characters

video-enhancement-using-single-image-super-resolution

video-object-segmentation

Video object segmentation allows computer vision to identify objects as they move through the space in a given video. Here we presents One-Shot Video Object Segmentation (OSVOS), a CNN architecture to tackle the problem of semisupervised video object segmentation, that is, the classification of all pixels of a video sequence into background and foreground,given the manual annotation of one of its frames.

video-object-segmentation-from-one-frame-annotation

video-segmentation

Video Segmentation -which means segmentation of all objects of the scene and classifying them based on certain classes. • Video segmentation will have a major role in day to day activities like traffic counting ,movie editing ,individual tracking etc.

video-super-resolution

Nowadays the crime rates all over the globe are increasing at an alarming rate and for that measures are being taken but not quite so good. We witness these criminals are let loose everyday just because the evidence isn’t enough or isn’t strong enough to be held in the court of law. One of the major issues in gathering evidence for such heinous crimes that are captured in the CCTV camera is that its not an evidence good enough to be accepted by the court and the sole reason behind it being the quality of these videos are not good enough in terms of resolution and which aches our hearts out to see those criminals living the life of free man without any guilt or punishment. This is why we interns thought we might put our knowledge about deep learning into some good use for the society and help to maintain the valuable parameters of the jurisdiction to some extent.

video-surveillance-system-to-identify-and-track-the-target-person

—Video Surveillance System is a used for preventing abnormal activity of the persons which leads to crime, loss of property or loss of lives. The abnormal persons are targeted by their activity and tracked across multiple frames to ensure the security of the public. This can be done effectively by our project of Video Surveillance System to identify and Track the Target person using Multi-Task Mask RCNN.

violence-detection-using-convolutional-lstm-deployed-over-drone

violence-recognition

violent-individuals-identification

With the increase in crime rate, smart surveillance systems are the need of the hour.

visual-relationship-detection-using-vtranse-network-paper

Visual relationships for e.g. person is talking to another person and a clock above the person offer a comprehensive scene. Inspired by the recent advances in relational representation learning of knowledge bases and convolutional object detection networks, we used a Visual Translation Embedding network (VTransE) for visual relation.

voice-face-and-action-interaction-model-of-nao-humanoid-robot

This project describes the interactive functionalities of NAO humanoid robot with human activities. It demonstrates the face recognition, voice recognition and various human activities using sensors. The above tasks are carried out in Choregraphe based Graphical User Interface (GUI) module.

voip-spam-detection-using-machine-learning

volume-control-using-hand-gestures-recognition

Gesture recognition helps computers to understand human body language. This helps to build a more potent link between humans and machines, rather than just the basic text user interfaces or graphical user interfaces (GUIs). In this project for gesture recognition, the human body's motions are read by computer camera. The computer then makes use of this data as input to handle applications. The objective of this project is to develop an interface which will capture human hand gesture dynamically and will control the volume level. For this, Deep Learning techniques such as Yolo model, Inception Net model+LSTM, 3-D CNN+LSTM and Time Distributed CNN+LSTM have been studied to compare the results of hand detection. The results of Yolo model outperform the other three models. The models were trained using Kaggle and 20% of the videos available in 20 billion jester dataset. After the hand detection in captured frames, the next step is to control the system volume depending on direction of hand movement. The hand movement direction is determined by generating and locating the bounding box on the detected hand.

wake-up-word-detection

Wake-up-word(WUW)system is an emerging development in recent times. Voice interaction with systems have made life ease and aids in multi-tasking. Apple, Google, Microsoft, Amazon have developed a custom wake-word engine, which are addressed by words such as ‘Hey Siri’. ‘Ok Google’, ‘Cortana’, ‘Alexa’. Our project focuses initially only detection and response to a customized wake-up command. The wake-up command used is “GOLUMOLU”. A wake-up-word detection system search for specific word and reads the word, where it rejects all other words, phrases and sounds. WUW system needs only less memory space, low computational cost and high precision. Artificial Neural Networks(ANN) have reduced the complexity, computational time, latency, thus the efficiency of system has improved. Deep learning has improved the efficiency of automatic speech recognition(SR), where wake word detection is a subset of SR but unlike keyword spotting and voice recognition. A deep learning RNN model is used for the training of the network. RNN are specifically used in case of temporal sequence data and has the ability to process data of different length but of same dimension. For training a model, labelled dataset is needed. We generated three forms of data: golumolu, negative and background. Such that, the model learns circumspectly and attentively detects when specific word found. To start communication with system, the wake word should be delivered. The main task of WUW detection system is to detect the speech, to identify WUW words among spoken words, to check whether the word spoken in altering context.

water-anomaly-detection-using-autoencoder

water-anomaly-detection-wad-autoencoder

The identification of water that is fit for consumption and usage has proven to be a challenge. The parameters of water that are analyzed are typically Temperature, pH, turbidity, SAC etc.To develop an event detector to accurately predict changes in a time series of drinking water composition data, we selected the AE and RBM .

water-anomaly-detection-wad-using-autoencoder

weapon-detection-with-heatmap

weed-detection-in-dense-culture-using-deep-learning-

In recent years, Weeds have been responsible for the agricultural losses. To get rid of this problem the farmers have to uniformly spray the whole field with the weedicides which requires a huge quantity of weedicides as well as is affects the environment.

wider-face-and-persons-challenge-2019-track-2-pedestrian-detection

Pedestrian Detection is an application of computer vision which is close to object detection, which has a wide range of applications. It can be used in surveillance monitoring, autonomous vehicles, face recognition, etc. This project is based on pedestrian detection that detects pedestrians and cyclists equally. The detector used must have high accuracy and precision. The model used is Faster R-CNN, which has more accurate than YOLOv3 but is comparatively slower.

wider-face-detection

Face detection and alignment in unconstrained environment are challenging due to various poses, illuminations and occlusions. We propose a deep cascaded multi-task framework which boost up the detection performance. In particular, our framework leverages a cascaded architecture with three stages of carefully designed deep convolutional networks to predict face and landmark location in a coarse-to-fine manner. Our method achieves superior accuracy over the state-of-the-art techniques on the challenging WIDER FACE benchmarks for face detection while keeps real time performance.

wider-face-detection-