# Submissions for: Automatically Segment and Label Objects in Video
|#| Submission | Links |
|-|-|-------------------------------------------|
|1|[Hybrid Multi-Class Video Segmentation with Deep Learning](https://github.com/MohitGoyal09/MATLAB_CHALLENGE_203)
This hybrid multi-class system uses YOLOv4 detection and a multi-stage classical vision pipeline for mask refinement, automatically generating high-quality pixel labels for video objects and reducing manual labeling time by 85.7%.
**Authors:** Angel Gupta, Mohit Goyal
**Affiliation:** bharati vidyapeeth college of engineering ,GGSIPU
**Submission Date:** 2025-08-20|![]()
[](https://github.com/MohitGoyal09/MATLAB_CHALLENGE_203)
[](https://matlab.mathworks.com/open/github/v1?repo=MohitGoyal09/MATLAB_CHALLENGE_203)
[](https://github.com/mathworks/MathWorks-Excellence-in-Innovation/blob/main/megatrends/Artificial%20Intelligence.md) [](https://github.com/mathworks/MathWorks-Excellence-in-Innovation/blob/main/megatrends/Computer%20Vision.md)|
|2|[Advanced Video Object Segmentation for MATLAB Labeler](https://github.com/Madhumitha282002/MATLAB_203/)
An advanced multi-technique video object segmentation algorithm for MATLAB's Video Labeler, leveraging optical flow, appearance modeling, and superpixel regularization to achieve efficient, accurate, and automated video annotation.
**Authors:** Madhumitha Katam
**Affiliation:** Arizona State University
**Submission Date:** 2025-07-31|![]()
[](https://github.com/Madhumitha282002/MATLAB_203/)
[](https://matlab.mathworks.com/open/github/v1?repo=Madhumitha282002/MATLAB_203)
[](https://github.com/mathworks/MathWorks-Excellence-in-Innovation/blob/main/megatrends/Artificial%20Intelligence.md) [](https://github.com/mathworks/MathWorks-Excellence-in-Innovation/blob/main/megatrends/Computer%20Vision.md)|