@bluevalhalla those are indeed only genera. And yes, I also found that there are some crucial genera missing. Nevertheless might it still be handy for you to make a rough separation before doing any human verification. 

Regarding the video processing: I didn't catch that you're looking for an MD results file that does all frames. MD creates that, but EcoAssist deletes it again (to avoid confusion with the video-level detection file). If you want the frame-level detection file, just out comment these lines. Then you'll have both. 

Answering Peter's question to me above, about how MegaDetector's video processing stuff decides which detection to include in the video-level output file:

• By default, the function that matters just chooses the highest-confidence detection for each category (person/animal/vehicle) as the "canonical detection" for that video (for posterity, that happens here).
• The frame number corresponding to that detection was not previously included in the output (since 99.99% of users would have no way of using this, but it sounds like you're in the other 0.01% :) ), so Peter just submitted a PR to the MegaDetector repo to include this in the output for a video; I'll merge that PR today, so I expect you'll see this in a forthcoming version of EcoAssist.

But ditto what Peter said, it sounds like you actually want the frame-level output file, and Peter gave you a tip about how to get that.

Also FWIW you will almost never want to process every frame of a video, nor will you likely want to train your models on every frame of a video; you're not getting a lot of new information from adjacent frames.  Typically when MegaDetector users process videos, I advise sampling videos down to ~3fps, which is usually every 10th frame.

Actually my Raspberry Pi application is a sound localizer not related to image recognition. My image recognition related project runs on Jetsons and higher.

But I think recognizing bugs on a drone would likely be challenging. You would have to have sufficient detail to get good recognition which would be a very narrow field of view and then vibration also becomes an issue.

For example, the trainings on just the coco dataset seems to distill the recognition of people to a multi-segmented thing with bits sticking out. So spiders on camera lens are highly likely to be seen as people. To get better results much more training data is needed. I expect it's also likely to be the case for insects, really large amounts of training data would be needed to tell the difference between different types.

Hello @ThomasGray_Argos ,

I need to add transmitting component, so please provide me the list of these. Actually, I don't know much about it. I am interested in learning new things.

I am new here but a real enthusiast and loving this community so far. I have a background in technology and feel I could help with documentation, at least for starters. If you have any other question apart from this, you can ask. I will try my best to solve your problem. 

That’s not a lot of power, which is great!

Welcome, @Dan_Ford ! I'm Alex from the WILDLABS community team. We're so happy to have you here. 

I definitely recommend sharing that job opening on WILDLABS as a career opportunity, that way it will show up in our biweekly digest and get seen by more people! To do that, just click +Post in the top right corner, then click "Careers" and fill in all the relevant information! We'd be happy to share the opening on our socials as well. 

I've also gone ahead and created a page for Landscape on The Inventory, our wiki-style database of conservation tech information. It would be great if you could take a look and add any missing information--I've gone ahead and given you editing access.  Also, you can connect your WILDLABS profile to Landscape's page so that people know to contact you with any questions about the organisation! To do this, go to your profile and select “Settings” on your side-bar, and then select the “About you” tab. Search and select your organisation name in the “Organisations(s)” section and click save.

Let me know if you have any questions :) 

Hi David - have you seen this opportunity? It may be a good one to apply to for support (the Darwin Initiative). 

Thanks for your insights Kim!

But the use of the browser password manager should make it no big deal for people even if they got logged out once a day. Which I’m sure is not the case for the wildlabs site.'

"Big deal" is very subjective...

I use firefox on linux and have tabs unloaded when not in use for ~30 minutes. After 24-48 hrs if I click on the wildlabs tab I'm logged out. I then have to click on login and have my password manager fill out the form. Then it brings me to my profile page (why? why doesn't it go back to where I was which is where I wanted to be, just logged in?), then I have to click on the feed page, which displays the global feed which is also not of my interest. Finally I get to click on my feed. Is is a "big deal"? I can't claim it is. Does it keep me from checking wildlabs? Definitely!  I participate in a dozen forums/discords and wildlabs is the most annoying to check in for a quick "oh, let's see whether there's something interesting".

It's still a choice to close your tabs. Closing all your tabs tends to get you to have to login again.

However.... Despite the expiry date on the cookie. linkedin.com logs in without ever having to re-enter your password. So I guess there is a way in principle to make it so you don't have to log back in. Such would a way then also work with your auto tab closer.

However, that's not something that I'm going to investigate. If someone wanted to, the clue would be in the cookies and the final answer would be available from the wildlabs site developer. Myself, I would never have guessed that so many people would find this annoying. But everyone is different.

Hi Cathy! Please check out the below article on group managers. Essentially, we're asking for a 12 month commitment of 1 hour a week (more or less) to promote engagement in WILDLABS groups by sparking conversations, engaging with people's content, and planning 1 virtual event a month to bring the community together. (This could be bringing a speaker in, or just having a 45 minute coffee call for people in the sector to get to know each other and share their work.)

As a group manager, you have the full support of the WILDLABS community team. We have a dedicated Slack channel, monthly calls to support you in your group management, and so much more.

Being a group manager is a great way to give back to the conservation tech community, steer conversations in the direction you think is important, foster a vibrant space for collaboration, and build your network. Shoot me an email if you're interested ([email protected])! I'm happy to hop on a call to discuss more in-depth. 

Unfortunately I may fail to deliver so I cannot take it up. Am off net at times for two weeks straight. 

But excited about everything GIS so I follow keenly. 

Thank you for the explanation. 🤍