Sovereign 8s.is mirror of standardgalactic/Pseudo-Depth-Detection
Find a file
Repository files (latest commit first)
Filename Latest commit message Latest commit date
2026-01-23 18:47:43 +08:00
code commit: Add files 2026-01-23 18:47:43 +08:00
data commit: Add files 2026-01-23 18:47:43 +08:00
README.md commit: Add files 2026-01-23 18:47:43 +08:00

Pseudo-Depth-Based object detection

This repository provides code and data for the study on "Pseudo-depth-based deep neural network model for object detection".

Dataset The experiment was trained and validated using the M3FD dataset and a subset of COCO. Both the M3FD dataset and COCO dataset are publicly available. The M3FD dataset is available for download via the link below: https://github.com/JinyuanLiu-CV/TarDAL The COCO dataset is available for download via the link below: https://cocodataset.org/ The list of image IDs used in the experiment can be found in the data/ids directory. The depth images used in the experiment and the weights generated from the experiment are available for download via the link below: https://pan.xunlei.com/s/VOjdypxPdKJdgCNyPgwu24lXA1# Extract code:htyj

The structure of an example dataset folder is as follows:

train/val
├── rgb
│   └── images
│       ├── xxx.jpg
│       ├── xxx.jpg
├── depth
│   └── images
│       ├── xxx.jpg
│       ├── xxx.jpg
│   ├── labels
│   │   ├── xxx.txt
│   │   ├── xxx.txt

Usage

  1. Download the required data set and move to folder./dataset.
  2. You need to download the YOLOv11 source code and replace the files in the code directory with the appropriate location to run the program.