A video scene detection algorithm is designed to detect a variety of different scenes within a video

Last update: Jan 04, 2022

Overview

Scene-Change-Detection

The detection of scenes change is a simple problem that human beings face, but it gets much harder to handle autonomously a device that generally includes complex calculations and algorithms.

A video scene detection algorithm is designed to detect a variety of different scenes within a video. There is a very simple definition for a scene: It is a series of logically and chronologically related shots taken in a specific order to depict an over-arching concept or story. The identification of video scenes, in many video analysis applications, is a crucial pre-processing step. A dataset for video scene detection known as the Open Video Scene Detection (OVSD) dataset has been provided in order to evaluate algorithms for video scene detection. Videos in the dataset have an open-source nature, which makes them an ideal product to be used by academics, as well as industry researchers alike.

DATASET

Dataset 2012 - In the dataset, there are six video categories, and in each category, there are four to six video sequences
IBM video Scene Change Detection
A dataset for video scene detection known as the Open Video Scene Detection (OVSD) dataset has been provided in order to evaluate algorithms for video scene detection.

MODEL

VGG16 was used for this project.VGG16 is a convolutional neural network model proposed by K. Simonyan and A. Zisserman from the University of Oxford in the paper “Very Deep Convolutional Networks for Large- Scale Image Recognition”. The model achieves 92.7% top-5 test accuracy in ImageNet, which is a dataset of over 14 million images belonging to 1000 classes.

LIBRARIES USED

Numpy for array manipulation
OpenCV (cv2) for Image Augmentation
Keras for building the Neural Network
Matplotlib for plotting visuals

COMPILATION USED

Loss function selected is sparse categorical cross-entropy
Optimizer selected is Adam
Validation metric chosen is accuracy

Training

No of epochs = 5
Batch size = 1

REPORT

https://drive.google.com/file/d/1cwoP5cRJ5D76PvHV_WjCDRoSJIf0h9du/view?usp=sharing

COLLABORATORS

Neel kumar arya and Ashish Vidyarthi

A video scene detection algorithm is designed to detect a variety of different scenes within a video

Related tags

Overview

Scene-Change-Detection

DATASET

MODEL

LIBRARIES USED

COMPILATION USED

Training

REPORT

COLLABORATORS

License

Owner

The King is Naked: on the Notion of Robustness for Natural Language Processing

Source code of AAAI 2022 paper "Towards End-to-End Image Compression and Analysis with Transformers".

SpineAI Bilsky Grading With Python

Python framework for Stochastic Differential Equations modeling

REBEL: Relation Extraction By End-to-end Language generation

"Learning and Analyzing Generation Order for Undirected Sequence Models" in Findings of EMNLP, 2021

NeurIPS-2021: Neural Auto-Curricula in Two-Player Zero-Sum Games.

Sequential GCN for Active Learning

FusionNet: A deep fully residual convolutional neural network for image segmentation in connectomics

Implements pytorch code for the Accelerated SGD algorithm.

PyTorch code for our ECCV 2018 paper "Image Super-Resolution Using Very Deep Residual Channel Attention Networks"

A Library for Modelling Probabilistic Hierarchical Graphical Models in PyTorch

Large-scale open domain KNOwledge grounded conVERsation system based on PaddlePaddle

Clockwork Variational Autoencoder

Yolo algorithm for detection + centroid tracker to track vehicles

A light and fast one class detection framework for edge devices. We provide face detector, head detector, pedestrian detector, vehicle detector......

Transfer SemanticKITTI labeles into other dataset/sensor formats.

Pytorch implementation of Decoupled Spatial-Temporal Transformer for Video Inpainting

Probabilistic Programming and Statistical Inference in PyTorch

PyTorch implementation of SampleRNN: An Unconditional End-to-End Neural Audio Generation Model