# Assignment 4 - COMP6970 - Fall 2024 - LATE — Data

来源：https://www.kaggle.com/competitions/assignment-4-comp-6970-fall-2024-late/data

本地 Meta Kaggle 快照；不是本次实时页面。

# Dataset Description

You will be predicting labels for each pixel in an input RGB image. There are 22 possible labels, including '0' for background and '255' for void/border.

## Folders

* ** dataset/JPEGImages/train/ ** - training set of RGB images.
* ** dataset/SegmentationClass/train/ ** - training set of per-pixel labels for each training RGB image.
* ** dataset/JPEGImages/val/ ** - validation set of RGB images. Accuracy on this set determines your position on the leaderboard.

## Labels

0 - Background
1 - Aeroplane
2 - Bicycle
3 - Bird
4 - Boat
5 - Bottle
6 - Bus
7 - Car
8 - Cat
9 - Chair
10 - Cow
11 - Diningtable
12 - Dog
13 - Horse
14 - Motorbike
15 - Person
16 - Pottedplant
17 - Sheep
18 - Sofa
19 - Train
20 - Tvmonitor
255 - Void/Object Border