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

banditml is a lightweight contextual bandit & reinforcement learning library designed to be used in production Python services.

Pytorch implementation of Integrating Tree Path in Transformer for Code Representation

Codebase for INVASE: Instance-wise Variable Selection - 2019 ICLR

FCAF3D: Fully Convolutional Anchor-Free 3D Object Detection

Multiple custom object count and detection using YOLOv3-Tiny method

custom pytorch implementation of MoCo v3

Official repository for the paper "GN-Transformer: Fusing AST and Source Code information in Graph Networks".

This project deploys a yolo fastest model in the form of tflite on raspberry 3b+. The model is from another repository of mine called -Trash-Classification-Car

A Rao-Blackwellized Particle Filter for 6D Object Pose Tracking

FCOSR: A Simple Anchor-free Rotated Detector for Aerial Object Detection

DeepLabv3+：Encoder-Decoder with Atrous Separable Convolution语义分割模型在tensorflow2当中的实现

code for Multi-scale Matching Networks for Semantic Correspondence, ICCV

Implementation of Convolutional LSTM in PyTorch.

Materials for my scikit-learn tutorial

The official implementation of the IEEE S&P`22 paper "SoK: How Robust is Deep Neural Network Image Classification Watermarking".

GRaNDPapA: Generator of Rad Names from Decent Paper Acronyms

Create and implement a deep learning library from scratch.

Optimus: the first large-scale pre-trained VAE language model

Semantic segmentation models, datasets and losses implemented in PyTorch.

Custom IMDB Dataset is extracted between 2020-2021 and custom distilBERT model is trained for movie success probability prediction