LAVIS VS AndroidTacticalAssaultKit-CIV

For multimodal, there's China (*many), then Salesforce.

Been trying for all day to get a working inference for this example: https://github.com/salesforce/LAVIS/tree/main/projects/instructblip

There's also been big breakthroughs in computer vision. Not that long ago it was hard to recognize if a photo contained a bird; that's solved now by models like CLIP, Yolo, or Segment Anything. Now research has moved on to generating 3D scenes from images or interactively answering questions about images.

The new hotness is multimodal vision-language models like InstructBLIP that can interactively answer questions about images. Check out the examples in the github repo, I would not have thought this was possible a few years ago.

Salesforce’s BLIP family has a new member– InstructBLIP, a vision-language instruction-tuning framework using BLIP-2 models. It has achieved state-of-the-art zero-shot generalization performance on a wide range of vision-language tasks, substantially outperforming BLIP-2 and Flamingo. (Source)

Github

Most of my workflows are self-made. For captioning I used Blip-2 in a custom script I made that automates the process by going into directories and their sub-directories and creates a .txt file beside each image. This way I can keep my images organized in their proper directories, without having to put dump them all in a single place.

I haven't had that issue before, but this github issues link might help you. Note: I'm new to TAK development so sorry if this isn't helpful. Linky

I can see the plugin SDK for 4.6 here Releases · deptofdefense/AndroidTacticalAssaultKit-CIV · GitHub but if I use that to build for TAK version 4.8.1 on the 3rd party pipeline, the Plugin won't load into CivTAK 4.8.1, despite being listed as current. I built the plugin with the 4.6 SDK and set ext.ATAK_VERSION = "4.8.1" in the app gradle file.

Without access to military gear I know of guys building ATAK https://github.com/deptofdefense/AndroidTacticalAssaultKit-CIV with specialised APIs that support rural Fire services etc . And was thinking that a drone with a FLIR camera that could communicate to the ATAK ground station via a TCP/UDP carrier would be viable. There's also our beach life savers that use drones to surveillance for sharks and rips... I wonder if you had that imagery that could use AI like facial recognition to quickly identify hazards.

I'm currently trying to dip my feet into the world of civtak plugin development and have been running into a few issues getting the latest version (SDK version 4.5.1.13) of the sdk to build using gradle on android studio. When gradle tries to build the plugin template project, I keep getting this particular error.

I've been working through the available dev guidance to build my first plugin. I started with Ballantyne's 'Build an ATAK Dev Environment' but I haven't been able to get the project to build successfully. I keep getting a Conan install error. Is this a necessary step for ATAK-CIV plugin development?

Actually, it is open source, since 2020.

Link to the source code, in case you didn't already have it: https://github.com/deptofdefense/AndroidTacticalAssaultKit-CIV