Forum - Recent Threads

I made a Lightweight Fingers Detection model That dosent uses any Training Data or any AI and is almost accurate

Aniket_kumar_raj — Fri, 27 Feb 2026 15:13:50 GMT

Do you know where was i For a few Days....

I was Grounded home with no internet so i made something that i was not able to find on internet... but first lets start from beginning to explain everything .look at this thing

[View:https://www.youtube.com/shorts/TsgXpnmGRMQ:640:360]

I saw this short in my Short feed and tried to create it but the one thing required was a lightest hand detection model, but none ai could make it and i wasn't able to find any on internet , then i took everything in my own hands and made this code that detects fingers smoothly, but dont move your face in box, only hands are allowed in the green box

here is the code that i made roasting my 1.3 megapixel camera of 17 years old laptop that i use for daily study..

to use this you must have these liblaries in python --->

cv2 , numpy , math

I did add comments so its easy to understand and variables are good too

[embed:dc8ab71f-3b98-42d9-b0f6-e21e02a0f8e2:f87614d8-e258-43da-8251-f2f6a9309a8e:type=python&text=import%20cv2%0Aimport%20numpy%20as%20np%0Aimport%20math%0A%0A%0A%0A%23%20This%20function%20does%20literally%20nothing%2C%20like%20my%20productivity%20on%20a%20Friday%0Adef%20hole_in_the_space_time_continuum%28x%29%3A%20pass%0A%0A%0A%0A%23%20Waking%20up%20the%201.3MP%20beast%0Aeye_of_sauron%20%3D%20cv2.VideoCapture%280%29%0Acv2.namedWindow%28%22The_Lag_Machine%22%29%0A%0A%0A%0A%23%20Trackbars%20for%20when%20the%20lighting%20in%20your%20room%20changes%20by%201%25%0Acv2.createTrackbar%28%22Magic_Slider%22%2C%20%22The_Lag_Machine%22%2C%20203%2C%20255%2C%20hole_in_the_space_time_continuum%29%0Acv2.createTrackbar%28%22Jitter_Control%22%2C%20%22The_Lag_Machine%22%2C%2015%2C%2050%2C%20hole_in_the_space_time_continuum%29%0A%0Awhile%20True%3A%0A%23%20Trying%20to%20grab%20a%20frame%20before%20the%20laptop%20overheats%0Ais_it_alive%2C%20blurry_mess%20%3D%20eye_of_sauron.read%28%29%0Aif%20not%20is_it_alive%3A%20break%0A%0A%0A%0A%23%201.%20Flip%20it%20because%20reality%20is%20disappointing%0Ablurry_mess%20%3D%20cv2.flip%28blurry_mess%2C%200%29%0Athe_forbidden_zone%20%3D%20blurry_mess%5B50%3A450%2C%2050%3A450%5D%0Acv2.rectangle%28blurry_mess%2C%20%2850%2C%2050%29%2C%20%28450%2C%20450%29%2C%20%280%2C%20255%2C%200%29%2C%202%29%0A%0A%0A%0A%23%202.%20Add%20filters%20to%20hide%20the%202009%20sensor%20quality%0Adepression_mode%20%3D%20cv2.cvtColor%28the_forbidden_zone%2C%20cv2.COLOR_BGR2GRAY%29%0Asmudge_tool%20%3D%20cv2.GaussianBlur%28depression_mode%2C%20%2815%2C%2015%29%2C%200%29%0A%0A%0A%0Acurrent_mood%20%3D%20cv2.getTrackbarPos%28%22Magic_Slider%22%2C%20%22The_Lag_Machine%22%29%0A_%2C%20crunchy_pixels%20%3D%20cv2.threshold%28smudge_tool%2C%20current_mood%2C%20255%2C%20cv2.THRESH_BINARY_INV%29%0A%0A%23%20Trying%20to%20glue%20the%20pixels%20back%20together%0Aduct_tape%20%3D%20np.ones%28%285%2C5%29%2C%20np.uint8%29%0Acrunchy_pixels%20%3D%20cv2.morphologyEx%28crunchy_pixels%2C%20cv2.MORPH_CLOSE%2C%20duct_tape%29%0A%0Awiggly_lines%2C%20_%20%3D%20cv2.findContours%28crunchy_pixels%2C%20cv2.RETR_EXTERNAL%2C%20cv2.CHAIN_APPROX_SIMPLE%29%0A%0Achicken_nuggets%20%3D%20%5B%5D%0Alies_on_screen%20%3D%20%22Looking%20for%20signs%20of%20life...%22%0A%0Aif%20wiggly_lines%3A%0Athe_big_blob%20%3D%20max%28wiggly_lines%2C%20key%3Dcv2.contourArea%29%0Aif%20cv2.contourArea%28the_big_blob%29%20%3E%202500%3A%0A%0A%0A%0A%23%20Finding%20the%20center%20of%20the%20mess%0Asecret_map%20%3D%20cv2.distanceTransform%28crunchy_pixels%2C%20cv2.DIST_L2%2C%205%29%0A_%2C%20max_stress%2C%20_%2C%20the_pixel_god%20%3D%20cv2.minMaxLoc%28secret_map%29%0Aholy_x%2C%20holy_y%20%3D%20int%28the_pixel_god%5B0%5D%29%2C%20int%28the_pixel_god%5B1%5D%29%0A%0A%23%20Checking%20if%20it%27s%20a%20hand%20or%20just%20a%20ghost%0Aprotective_shell%20%3D%20cv2.convexHull%28the_big_blob%29%0Ahow_chunky%20%3D%20float%28cv2.contourArea%28the_big_blob%29%29%20%2F%20cv2.contourArea%28protective_shell%29%0A%0Aif%20how_chunky%20%3C%200.88%3A%0Achaos_factor%20%3D%20cv2.getTrackbarPos%28%22Jitter_Control%22%2C%20%22The_Lag_Machine%22%29%0Amath_points%20%3D%20the_big_blob.reshape%28-1%2C%202%29%0Apoint_count%20%3D%20len%28math_points%29%0A%0Afor%20i%20in%20range%280%2C%20point_count%2C%202%29%3A%0A%23%20Geometry%20is%20harder%20than%20my%20life%20choices%0Aa%2C%20b%2C%20c%20%3D%20math_points%5B%28i-chaos_factor%29%25point_count%5D%2C%20math_points%5Bi%5D%2C%20math_points%5B%28i%2Bchaos_factor%29%25point_count%5D%0Avector_1%2C%20vector_2%20%3D%20a%20-%20b%2C%20c%20-%20b%0Alen_1%2C%20len_2%20%3D%20np.linalg.norm%28vector_1%29%2C%20np.linalg.norm%28vector_2%29%0A%0Aif%20len_1%20%3E%200%20and%20len_2%20%3E%200%3A%0Aconfusing_angle%20%3D%20math.degrees%28math.acos%28np.clip%28np.dot%28vector_1%2Cvector_2%29%2F%28len_1%2Alen_2%29%2C%20-1.0%2C%201.0%29%29%29%0A%0A%23%20Is%20it%20a%20finger%3F%20Or%20just%20a%20glitch%20in%20the%20Matrix%3F%0Adist_to_shell%20%3D%20abs%28cv2.pointPolygonTest%28protective_shell%2C%20%28float%28b%5B0%5D%29%2C%20float%28b%5B1%5D%29%29%2C%20True%29%29%0Adist_to_ego%20%3D%20np.linalg.norm%28b%20-%20np.array%28%5Bholy_x%2C%20holy_y%5D%29%29%0A%0Aif%2010%20%3C%20confusing_angle%20%3C%2045%20and%20b%5B1%5D%20%3C%20%28holy_y%20%2B%2050%29%20and%20dist_to_shell%20%3C%208%3A%0Aif%20dist_to_ego%20%3E%20%28max_stress%20%2A%201.4%29%3A%0Aif%20not%20any%28np.linalg.norm%28b%20-%20np.array%28prev%29%29%20%3C%2030%20for%20prev%20in%20chicken_nuggets%29%3A%0Achicken_nuggets.append%28tuple%28b%29%29%0A%0Alies_on_screen%20%3D%20f%22FINGERS%20OR%20NUGGETS%3A%20%7Blen%28chicken_nuggets%29%7D%22%0Aelse%3A%0Alies_on_screen%20%3D%20%22ANGRY%20FIST%22%0A%0A%23%20Draw%20green%20circles%20so%20we%20feel%20like%20hackers%0Acv2.circle%28the_forbidden_zone%2C%20%28holy_x%2C%20holy_y%29%2C%20int%28max_stress%29%2C%20%280%2C%20255%2C%200%29%2C%201%29%0Afor%20nugget%20in%20chicken_nuggets%3A%0Acv2.circle%28the_forbidden_zone%2C%20nugget%2C%2010%2C%20%28255%2C%200%2C%200%29%2C%20-1%29%0Acv2.line%28the_forbidden_zone%2C%20%28holy_x%2C%20holy_y%29%2C%20nugget%2C%20%280%2C%20255%2C%20255%29%2C%201%29%0A%0Acv2.putText%28blurry_mess%2C%20lies_on_screen%2C%20%2860%2C%2040%29%2C%201%2C%202%2C%20%280%2C%20255%2C%200%29%2C%202%29%0Acv2.imshow%28%22The_Lag_Machine%22%2C%20blurry_mess%29%0Acv2.imshow%28%22What_The_Computer_Sees%22%2C%20crunchy_pixels%29%0A%0A%0A%0A%0A%0A%23%20Press%20%27q%27%20to%20escape%20this%20nightmare%0Aif%20cv2.waitKey%281%29%20%26%200xFF%20%3D%3D%20ord%28%27q%27%29%3A%20break%0A%0Aeye_of_sauron.release%28%29%0Acv2.destroyAllWindows%28%29]

RE: I made a Lightweight Fingers Detection model That dosent uses any Training Data or any AI and is almost accurate

Christopher678 — Tue, 03 Mar 2026 22:26:01 GMT

Great post

Is the next leap of electronics brain organoids?

stanto — Sat, 28 Feb 2026 16:39:09 GMT

There's not a loud majority talking about this tech' at the moment, but it's somewhat concerning. Using stem cells groups are "finally" combining electronics and organics, notably stem cells. Doing everything from running Doom (of course) to running LLMs on/with them. This should be raising all sorts of questions.

Gabriel Torch does a great job at summarising legit research papers on this topic.

[View:https://www.youtube.com/watch?v=uQEg3pg7w9E:640:360]

https://www.sciencedirect.com/science/article/pii/S0896627322010339

https://www.nytimes.com/2019/08/29/science/organoids-brain-alysson-muotri.html

https://www.newscientist.com/article/2517389-human-brain-cells-on-a-chip-learned-to-play-doom-in-a-week/

And a few more scary videos on organoids that are worth watching:

[View:https://www.youtube.com/watch?v=9QXDdjp3OqA:640:360]

[View:https://www.youtube.com/watch?v=bzMjTiY7o7g:640:360]

You can also, apparently, play with them yourself:
[View:https://www.youtube.com/watch?v=vg2Y82rYnHM:640:360]

Anyone else welcoming our new stem-cell-organoid overlords?

RE: Is the next leap of electronics brain organoids?

stanto — Sun, 01 Mar 2026 22:29:17 GMT

Gabriel Torch did a video where they covered that the organoids were 'dying' because.. well they were starved of oxygen and nutrients, which is now a solved problem apparently.

[quote userid="220772" url="~/technologies/ai-machine-learning/f/forum/56726/is-the-next-leap-of-electronics-brain-organoids/234112"]I'm not sure what the moral concern is[/quote]

And then the question comes up about sentience, and running them in a 'virtual environment' so that they 'don't get bored'.

RE: Is the next leap of electronics brain organoids?

stanto — Sun, 01 Mar 2026 22:26:49 GMT

[quote userid="328965" url="~/technologies/ai-machine-learning/f/forum/56726/is-the-next-leap-of-electronics-brain-organoids/234109"]Wow, this kind of caught me off guard.[/quote]

Same. It's really being kept 'on the down low'.

RE: Is the next leap of electronics brain organoids?

kmikemoo — Sun, 01 Mar 2026 22:01:04 GMT

Awesome. Couple this with the blind trust of AI and we'll be living a video game. Or a movie. Not excited.

RE: Is the next leap of electronics brain organoids?

robogary — Sat, 28 Feb 2026 21:23:18 GMT

RE: I made a Lightweight Fingers Detection model That dosent uses any Training Data or any AI and is almost accurate

DAB — Sat, 28 Feb 2026 19:41:26 GMT

Very nice post.

RE: Is the next leap of electronics brain organoids?

DAB — Sat, 28 Feb 2026 19:24:30 GMT

This is a technology that could easily be miss used.

I understand the interest, but this could lead to some very bad results.

RE: Is the next leap of electronics brain organoids?

bradfordmiller — Sat, 28 Feb 2026 18:09:01 GMT

I'm not sure what the moral concern is, one doesn't need to use human brain cells. I had hoped to get into this area about a decade ago for robotics applications but we quickly figured out we didn't have the staff needed to pursue. The main point is that biological brains are much lower power and require just a few layers of neurons to be quite competent, particularly in physical tasks (balance, locomotion, hand/eye coordination, etc.)

RE: Is the next leap of electronics brain organoids?

beacon_dave — Sat, 28 Feb 2026 18:02:47 GMT

Last week I was listening to a podcast from back in in 2022 when the brain cells were starting to get good at playing Pong.

"Pong was a very basic video game developed in the 1970s, now Australian researchers have trained human brain cells in a dish to play the game, Dr Brett Kagan from Cortical Labs explains why."

https://www.bbc.co.uk/programmes/m001cxxv

Lab-grown brain cells play video game Pong

https://www.bbc.co.uk/news/science-environment-63195653

Looks like they got bored and moved onto Doom like everyone else.

RE: Is the next leap of electronics brain organoids?

dang74 — Sat, 28 Feb 2026 16:51:43 GMT

Wow, this kind of caught me off guard. I guess it creates a real grey area between man and machine and I suppose lab grown brains suddenly become a commodity.

RE: I made a Lightweight Fingers Detection model That dosent uses any Training Data or any AI and is almost accurate

dang74 — Sat, 28 Feb 2026 01:17:55 GMT

[quote userid="518254" url="~/members-area/f/forum/56723/i-made-a-lightweight-fingers-detection-model-that-dosent-uses-any-training-data-or-any-ai-and-is-almost-accurate/234058"] It’s a safety feature. Without it, the chicken_nuggets start manifesting in physical reality[/quote]

'Manifested Reality' I think that was one of the original dipping sauces available at McDonald's.

RE: I made a Lightweight Fingers Detection model That dosent uses any Training Data or any AI and is almost accurate

dougw — Fri, 27 Feb 2026 23:05:32 GMT

Here are some fingers...

RE: I made a Lightweight Fingers Detection model That dosent uses any Training Data or any AI and is almost accurate

Aniket_kumar_raj — Fri, 27 Feb 2026 16:35:49 GMT

i heard 90% of the internet is held together by duct_tape and hope_and_prayers. I’m just following industry standards.

RE: I made a Lightweight Fingers Detection model That dosent uses any Training Data or any AI and is almost accurate

Aniket_kumar_raj — Fri, 27 Feb 2026 16:32:06 GMT

The duct_tape is there to keep the hole_in_the_space_time_continuum from expanding. It’s a safety feature. Without it, the chicken_nuggets start manifesting in physical reality, and my GPU can't handle that kind of rendering.

RE: I made a Lightweight Fingers Detection model That dosent uses any Training Data or any AI and is almost accurate

dang74 — Fri, 27 Feb 2026 16:27:47 GMT

Indeed not to mention a variable called duct_tape.

RE: I made a Lightweight Fingers Detection model That dosent uses any Training Data or any AI and is almost accurate

battlecoder — Fri, 27 Feb 2026 16:17:01 GMT

OpenCV is such a fun framework to work with. Love your code and comments btw. Wish more people were this whimsical when they write code.

RE: I made a Lightweight Fingers Detection model That dosent uses any Training Data or any AI and is almost accurate

robogary — Fri, 27 Feb 2026 15:54:18 GMT

Nice. Your comments and variable names are hilarious ! You had too much fun :-)

Computer network design: Replacing the coffee-break napkin discussion.

colporteur — Wed, 02 Jul 2025 22:08:27 GMT

I am looking for input from members in the E14 community that have computer network experience. The block diagram is the drawing of a small farming business network.

The network is one IP subnet spread across three physical locations. The subnet has about 50 IP consumed across the three areas. The edge router is in the office. Wireless point-to-point devices connect the seed-plant and residence to the office.

Inside the seed-plant and residence are routers that route the programmable logic controller and home devices, respectively. This routers will remain. What is happening behind the seed-plant router is expanding.

My plan is to replace the current edge router with a router capable of creating multiple networks. Four subnetworks to start: office, seed-plant, residence and wifi-guest-network that doesn't exist today. These networks will be interconnected and have internet access. The new edge router will isolate the the wireless guest-network and provide internet access only. I hope to combine the VPN access into the router and eliminate the VPN appliance.

One valuable resource I miss since retirement is coffee-break time with the team. It was not uncommon for someone to start a discussion on a napkin drawing. The napkin formed a centrepiece for group discussions.

I’m looking for that napkin discussion with members from the E14 Community. I encourage you to ask questions and provide feedback. Even if you don’t have computer networking experience, please join the discussion. Answering your questions may flush out some ideas I wasn’t thinking about.

After retiring, I took on the role of break-fix for the company that owns this network. Now that the network is stable, and the company is proposing expansion, I feel it is time to implement the network plan I created five years ago.

What are your thoughts?

RE: Computer network design: Replacing the coffee-break napkin discussion.

bradfordmiller — Thu, 03 Jul 2025 12:52:20 GMT

One of the main advantages of a VLAN (and you can google for a more comprehensive list), is security, in that you might want seed plant data going to the owner's home computer, but not to the TV (where who knows where it might get re-transmitted to). VLANs let you set up explicit rules for data sharing or isolation, and might replace most of what you are doing with the subordinate routers in your configuration simplifying (re)configuration. VLANs can generally (you'd have to check your specific router or switch) for either port-based or protocol based membership. I use port based so I can, for instance, make sure any data flowing from/to my MIL's TV and smart appliances have no interaction with my office machines (which have often dealt with IP under NDA, etc. particularly during COVID). So with a relatively straightforward physical setup, you can then software configure multiple virtual networks (which can appear invisible to other virtual networks) and modify them as needs change.

An advantage of smart switches (not just enabling VLANs or trunking), is remote management, e.g., if there's an issue, I can use the web interface to the switch and quickly determine, e.g., if a particular device is using it's port, blasting out too much data, etc. That saves time when debugging issues (they don't come up that often, but just looking at blinky lights on the switch can be a problem when, in my case, I've got about 10 switches in various locations around the house, and that doesn't always help with finding the root cause of congestion!). Many of those switches are small & dumb, but having a smart switch they connect to makes it much easier to get an overview of what's going on in the network.

RE: Computer network design: Replacing the coffee-break napkin discussion.

shabaz — Thu, 03 Jul 2025 04:59:37 GMT

Hi Sean,

For a small deployment you can get away with a lot, so it's not unusual to see 'strange' things since they can occur over time as needs change. And that's fine, if it's working and can be managed, although complications may occur as the business grows and their needs change. If you were starting from scratch, you'd make the design more future-proof in terms of maintainability especially, if a more standard layout was followed.

Generally, a typical design 'pattern' would be to consider the seed plant, and the residence, as 'small branches' of an org, and the office could be considered to be another branch or a HQ. Then, the normal design would be to have a router at each location, i.e. separate network at each site as I think you're planning to do. You can get routers with built-in switch ports, or you could get a separate switch for the sites, and they would hang off the router, rather than the router hanging off switches on sites. And then you could also optionally have VLANs if desired too, that would be supported very easily by this typical design pattern, amd then you'll find you could retire the current separate routers you've got at the two branches, since they could be replaced by VLAN functionality provided the switch supports it, i.e. one less piece of hardware to manage, since I can't see what benefit the router is providing versus if it were replaced by ports on the switch configured for a different VLAN, and then that's one less piece of equipment to configure/manage.

Regarding manufacturers, such as the ones bradfordmiller mentions, these are well-known manufacturers for equipment designed for business use, which (for many reasons) I'd trust, and Firewalla has a great team behind it (I've personally known a couple of the founders there). These might cost more than typical home routers, but I think still a reasonable cost. Some will have a few built-in switch ports, so you may still need to buy a switch per site too. Example of the type of router I'm thinking of: Cisco 900 or 1100 series perhaps. I don't know the Firewalla portfolio of products in detail, but as mentioned, I expect it to be good too.

RE: Computer network design: Replacing the coffee-break napkin discussion.

colporteur — Thu, 03 Jul 2025 00:40:45 GMT

The edge router will be a MikroTik RB4011iGS+5HacQ2HnD-IN. It is currently an AX11000 gaming router. The price was right (free) and I was desperate in need of a performance solution. I was discouraged from using the MikroTik it in favour of more friendly options. I'm still not confident I have made the right choice. It has little in the way of hand holding to configure it. I am on the task of configuring openvpn, a configuration it supports.

Only one of the switches supports VLANs. That is the unit in the office area. The network I took over in 2018 was a /8 subnet from an Apple router in the residence. It has morphed into what it is today. I'm on the cusp of making a big shift.

The extra router in the house was my attempt at isolating the myriad of consumer electronics that taxes the resources of the edge router. The edge router originally was in the residence. The network flowed from their. The seedplant was a modem configured to network to the seedplant over an abandoned phone line through an outbuilding to the residence.

The router in the seedplant, with no input from me, was a vendors solutions to isolate the PLC motor controllers and sensors in the seedplant and add a management system to the seedplant.

There is another network (well not really) a bunch of IP's that are assigned to some poorly implemented development systems in the seedplant. At some point I will need to beat them into submission.

I don't see advantages to VLANs at this time. At times seedplant traffic needs to get to the residence. The owners wants to examine systems. Please correct me if VLAN holds something I am missing. I examined the option before looking at routing and turned left at routing instead of right to VLAN. Maybe VLAN is the future. It would require some hardware to implement.

RE: Computer network design: Replacing the coffee-break napkin discussion.

bradfordmiller — Wed, 02 Jul 2025 22:37:06 GMT

So keep in mind I'm not a networking expert, but I do run a 1G/10G setup at home.

I'd consider getting a "security router" for your edge router - they usually have builtin VPN support. I have a Cisco, but I've heard good things about Firewalla so that might be my next option when the Cisco stops getting updated (they tend to drop SOHO routers after about 5 years; I'm on my second one now. The good thing about the router I have is it's "multi-wan" so I can have a backup connection to a second ISP if my primary drops, or I can configure splitting the bandwidth between them.)

I'd also suggest using smart switches for the Seed Plant AND the residence as that will allow you to set up VLANs with different rules, e.g. making sure nothing from the seed plant gets to the residence, but both can be seen by the office (if that's what you want). That will give you configuration flexibility for future topological changes, as it would just require rule changes on the smart routers.

Smart switches and routers can also connect to each other using trunking. For example I trunk multiple 1G ports between my 10G switch and my 1G switch to increase the bandwidth between them.

Question: why do you have additional routers hanging off the seed plant and residence switches? Normally that connection would be reversed (router first, switch to expand the number of LAN ports on the router).

RE: what is the most underrated use case of AI and Machine Learning you've come across?

dougw — Tue, 01 Jul 2025 17:31:00 GMT

We have an AI that can pick clothing sizes for you.