bencevans / camtrap-detector Goto Github PK
View Code? Open in Web Editor NEWDetect Animals, Humans and Vehicles in Camera Trap Imagery. Powered by MegaDetector v5.
Home Page: https://camtrap.net/detector
License: Apache License 2.0
Detect Animals, Humans and Vehicles in Camera Trap Imagery. Powered by MegaDetector v5.
Home Page: https://camtrap.net/detector
License: Apache License 2.0
If the hogwatch team is going to use it in practice for large surveys, perhaps it needs to support processing large sets of locations?
One option is processing each subdir separately and outputting a file for each (you could then have an option to skip these if the output file exists already), but I'm not sure what 'level' this should be at.
Some form of website that automatically picks the correct build given the user's machine.
I have written code to analyse JSON files produced by CamTrap Detector ( Windows 64 bit, CPU only, v 0.4.0) , EcoAssist and the Megadetector API at https://github.com/agentmorris/MegaDetector/tree/main/api/synchronous. I noticed that CamTrap Detector is much faster in operation than the other two but the histogram of detection confidences is vastly different from either when processing the same set of 1581 images, all 2048 x 1440 pixels. Is there an explanation for this?
Add an option to obscure (heavy blur or black box) bboxes of humans in image exports
Export with Paths relative to IMAGE_DIR.
Ability to restart the app (once you've processed one directory, it's not clear how to process another without restarting the app)
HH Team
Add Installation Instructions to README
An 'auto'/quite-mode option (in other words, specify the output csv/json, then press go and it'll output those files and close the app)
Although this was the case with a previous GUI and people were unsure if it had crashed or completed. Maybe keep some UI showing completion.
EXIF data is missing from image exports.
Notarize macOS applications to reduce the installation friction with macOS complaining it can't be verified.
Related issues:
xattr -d com.apple.quarantine /Applications/CamTrap\ Detector.app
Related documentation for implementing:
Ability to change classification thresholds
HH Team
Should run in parellal ... rayon
Also it might be nice of the boxes around the objects in the images were brighter so they are easier to see in the daylight.
-- Chris
I would recommend using the MDv5 training size of 1280x1280; it can make a significant difference in a number of cases. (There are very very very esoteric cases where 640x640 actually works better than the native input size, but those are rare).
Is there any way to change the color or square(to capture where the animal is within picture)? I am only seeing white but would be better if I can use another color. Thanks!
- @minorimatsuoka1258
Package the model file using Tauri's additional files mechanism rather than include_bytes! ... https://tauri.app/v1/guides/building/resources
E.g all images that contain both an animal AND a human.
Hello, why is it that on macOS, after compiling your code, I used the otool - L CamTrap Detector
and displayed the dependency /usr/local/opt/opencv/lib/libopencv_core. 407. dylib,...
? I know this is a dependency on my local opencv environment, but it will cause me to share it with other macOS users who have not installed opencv and will not be able to use it, but I downloaded your published CamTrap. Detector_0.4.0_x64. dmg
I used otool - L CamTrap Detecto
in the package, but it did not display the dependency 'libopencv_xxx. dylib'. Why is there a difference
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.