https://github.com/kyegomez/odin
SOTA Classification at scale for UAVs, Drones, and much more
https://github.com/kyegomez/odin
computer-vision multimodal multimodal-data multimodal-deep-learning swarm-intelligence
Last synced: 10 months ago
JSON representation
SOTA Classification at scale for UAVs, Drones, and much more
- Host: GitHub
- URL: https://github.com/kyegomez/odin
- Owner: kyegomez
- License: mit
- Created: 2023-08-31T20:17:42.000Z (almost 3 years ago)
- Default Branch: main
- Last Pushed: 2024-03-11T13:37:12.000Z (over 2 years ago)
- Last Synced: 2025-06-08T22:27:40.491Z (about 1 year ago)
- Topics: computer-vision, multimodal, multimodal-data, multimodal-deep-learning, swarm-intelligence
- Language: Python
- Homepage: https://discord.gg/qUtxnK2NMf
- Size: 211 KB
- Stars: 5
- Watchers: 2
- Forks: 0
- Open Issues: 4
-
Metadata Files:
- Readme: README.md
- Funding: .github/FUNDING.yml
- License: LICENSE
Awesome Lists containing this project
README
[](https://discord.gg/qUtxnK2NMf)
# Odin
Super Fast and super cheap object detection at massive scale in less than 10 lines of code!
# Appreciation
* Lucidrains
* Agorians
# Install
`pip install vodin`
# Usage
Here are three examples demonstrating the usage of the `Odin` class from your provided code:
**Example 1: Basic Usage**
```python
# Import the necessary modules and classes
from odin import Odin
# Initialize the Odin object with paths and thresholds
odin = Odin(
source_weights_path="yolo.weights",
source_video_path="input_video.mp4",
target_video_path="output_video.mp4",
confidence_threshold=0.3,
iou_threshold=0.7
)
# Run the object to process the video
odin.run()
```
**Example 2: Custom Parameters**
```python
# Import the necessary modules and classes
from odin import Odin
# Initialize the Odin object with custom parameters
odin = Odin(
source_weights_path="custom_yolo.weights",
source_video_path="input_video.mp4",
target_video_path="output_video.mp4",
confidence_threshold=0.5,
iou_threshold=0.6
)
# Run the object to process the video
odin.run()
```
**Example 3: Advanced Usage**
```python
# Import the necessary modules and classes
from odin import Odin
# Initialize the Odin object with paths and thresholds
odin = Odin(
source_weights_path="yolo.weights",
source_video_path="input_video.mp4",
target_video_path="output_video.mp4",
confidence_threshold=0.3,
iou_threshold=0.7
)
# Customize further configurations if needed
odin.tracker.set_max_distance(50)
odin.box_annotator.set_box_color((0, 255, 0))
odin.model.set_device("cuda")
# Run the object to process the video
odin.run()
```
# Architecture
* [Odin utilizes YoloV7, weights can be downloaded here](https://drive.google.com/file/d/1yEYFq1jCIpklofMMhuqQKwyTfvj1hLQ1/view)
# License
MIT