Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

overengineered the ktp data extraction using vlm (llama3.2-vision)

Awesome kid

9,669 subscribers

35,358 views • 1 year ago •via X (Twitter)

Science & Technology

Anya Rossi• Live Now

Private livecam show

11 Comments

Ali Mustofa1 year ago

kemarin habis nyobain fine tuning hehe

Zano1 year ago

WARNING: Your data is leaked!⚠️ If you're using transparent blockchains, you must realize that all your transactions will be linked to your name in less than a few years, thanks to chain analysis and AI! Privacy is not for criminals; it's for the people!

muhajaya 1 year ago

ini running di local mas?

ericc1 year ago

betul di local

Bukhori M. Aqid1 year ago

ini kalo KTPnya kebalik (upside down) OCR masih akurat ga?

ericc1 year ago

menarik, miss 1 field ternyata yaitu goldar yg posisinya agak di pinggir bikin klasifikasi sederhana dulu mungkin di awal pake isProperPosition? sekalian deteksi hal-hal yg unwanted wahaha

just zeey1 year ago

Ocr nota mas

Fahmi Sidik1 year ago

Keren mas

Saint laurent1 year ago

Menarik mas. Coba pake foto ktp yang laminatingnya udh burem lecek lecek 😁

Yudha Rebel Heart1 year ago

llama vision ini kayaknya oke juga ya?

ericc1 year ago

lumayan mas, ada yg lebih ringan lagi moondream2 bisa di local (via ollama) maupun cloud

Related Videos

A bit of data extraction from audio. Using #python and #houdini

A bit of data extraction from audio. Using #python and #houdini

ՖյՊՓի ԻսֆֆԸԼԼ ☀🤖

44,408 views • 2 months ago

DarkWeb Scraping DeepWebHarvester is a powerful tool designed to scrape onion sites on the dark web, enabling comprehensive crawling and extraction of valuable data. – A demo that illustrates how I extract data from an onion site ethically using the tool: #Cybersecurity

DarkWeb Scraping DeepWebHarvester is a powerful tool designed to scrape onion sites on the dark web, enabling comprehensive crawling and extraction of valuable data. – A demo that illustrates how I extract data from an onion site ethically using the tool: #Cybersecurity

Tech Enthusiast

18,787 views • 10 months ago

How can people truly monetize their own data using blockchain? In conversation with sachi.eth — Anna Kazlauskas shares her vision for a new data economy, powered by vana, where individuals own, control, and profit from their personal data.

How can people truly monetize their own data using blockchain? In conversation with sachi.eth — Anna Kazlauskas shares her vision for a new data economy, powered by vana, where individuals own, control, and profit from their personal data.

Web3Unchained Podcast 🎙️

40,223 views • 10 months ago

Robotics and CV! 👓 Computer vision gives the robots a sense of vision and additional intelligence. Pralines are packed by a delta robot using computer vision. Packing chocolate pralines into a box with the delta robot. Unlike most robots, this one has a fixed TCP, and the base moves in response to vision data. To adjust the base, the vision system determines exactly where the praline is located on the conveyor belt. 🍫 Using this application, the system knows where each praline should be placed based on its orientation. A true engineering showcase!

Robotics and CV! 👓 Computer vision gives the robots a sense of vision and additional intelligence. Pralines are packed by a delta robot using computer vision. Packing chocolate pralines into a box with the delta robot. Unlike most robots, this one has a fixed TCP, and the base moves in response to vision data. To adjust the base, the vision system determines exactly where the praline is located on the conveyor belt. 🍫 Using this application, the system knows where each praline should be placed based on its orientation. A true engineering showcase!

Lukas Ziegler

78,611 views • 1 year ago

“China is using TikTok to steal all of our data” The data:

“China is using TikTok to steal all of our data” The data:

Kenny Lay (Parody)

2,646,531 views • 2 years ago

I built an automated AI competitor analyis team with multi-agents. It has 3 AI agents working together as a team: 1. Data Extraction Agent using Firecrawl 2. Analyst Agent using OpenAI GPT-4o 3. Competitor Research Agent using Perplexity AI or Exa AI 100% Opensource code.

I built an automated AI competitor analyis team with multi-agents. It has 3 AI agents working together as a team: 1. Data Extraction Agent using Firecrawl 2. Analyst Agent using OpenAI GPT-4o 3. Competitor Research Agent using Perplexity AI or Exa AI 100% Opensource code.

Shubham Saboo

65,158 views • 1 year ago

Van Dijk's close range shot had an XG of 0.84 🤯 Watch Liverpool v PSG live using Prime Vision for a data-driven match experience 📊

Van Dijk's close range shot had an XG of 0.84 🤯 Watch Liverpool v PSG live using Prime Vision for a data-driven match experience 📊

Prime Video Sport UK

29,875 views • 2 months ago

We present VLM-3R: a Vision-Language Model capable of 3D spatial reasoning from monocular video, grounding visual cues, geometry, and camera motion. ✅ No depth sensor ✅ No pre-built 3D maps ✅ End-to-end spatial + temporal reasoning 🔗 Code & benchmark: #VLM #3DVision #LLMs

We present VLM-3R: a Vision-Language Model capable of 3D spatial reasoning from monocular video, grounding visual cues, geometry, and camera motion. ✅ No depth sensor ✅ No pre-built 3D maps ✅ End-to-end spatial + temporal reasoning 🔗 Code & benchmark: #VLM #3DVision #LLMs

Zhiwen(Aaron) Fan

14,895 views • 1 year ago

Gemini has always been my go-to LLM for document extraction - but accurate bounding boxes with agentic vision are a huge value-add in the latest Gemini models. Shown: extracting structured data from an oil & gas lease in Ector County Texas with the best bboxes I've seen yet.

Gemini has always been my go-to LLM for document extraction - but accurate bounding boxes with agentic vision are a huge value-add in the latest Gemini models. Shown: extracting structured data from an oil & gas lease in Ector County Texas with the best bboxes I've seen yet.

Kyle Walker

79,593 views • 3 months ago

I'm shocked by how easy it is to run PydanticAI with a model running logically. Here we have: ollama running llama3.2 locally, being used by Pydantic AI extracting structured data from an input.

I'm shocked by how easy it is to run PydanticAI with a model running logically. Here we have: ollama running llama3.2 locally, being used by Pydantic AI extracting structured data from an input.

Samuel Colvin

18,084 views • 1 year ago

The rill "KTP ngendi" ?

The rill "KTP ngendi" ?

Txt Dari Jogja

36,962 views • 1 year ago

What happens when VLMs meet 3D foundation models? See VLM-3R (CVPR 2026). VLM-3R links a vision-language model (e.g., Qwen) with 3D geometric foundation models (e.g., CUT3R) at metric scale. Given an uncalibrated video, it moves beyond pixels to perceive and reason in 3D space. Code (open source):

What happens when VLMs meet 3D foundation models? See VLM-3R (CVPR 2026). VLM-3R links a vision-language model (e.g., Qwen) with 3D geometric foundation models (e.g., CUT3R) at metric scale. Given an uncalibrated video, it moves beyond pixels to perceive and reason in 3D space. Code (open source):

Zhiwen(Aaron) Fan

10,595 views • 3 months ago

Streamer forgets he’s in real life while using the Vision Pro😭

Streamer forgets he’s in real life while using the Vision Pro😭

FadeHubb

8,114,368 views • 2 years ago

Big AI eats your data, gives nothing back. Ram explains why this extraction model is broken. OpenLedger flips the script. Transparent data. Verifiable contribution. Rewards that flow back to the source. The AI future isn’t privately owned. It’s collaborative 🐙

Big AI eats your data, gives nothing back. Ram explains why this extraction model is broken. OpenLedger flips the script. Transparent data. Verifiable contribution. Rewards that flow back to the source. The AI future isn’t privately owned. It’s collaborative 🐙

Openledger

54,642 views • 11 months ago

Tactile Diffusion generates synthetic tactile images from sim data, capturing the complex illumination of the gel deformation. This research from UW & Meta AI is the first method using diffusion to close the sim2real gap for vision-based tactile sensing. Read the paper ⬇️

Tactile Diffusion generates synthetic tactile images from sim data, capturing the complex illumination of the gel deformation. This research from UW & Meta AI is the first method using diffusion to close the sim2real gap for vision-based tactile sensing. Read the paper ⬇️

AI at Meta

100,151 views • 3 years ago

How Keeper is using GPT-4 for matchmaking. It takes profile data & preferences, determines if the match is worth pursuing & automates the followup. With computer vision for the physical, you can filter on anything and find your ideal partner.

Jake Kozloski

2,975,614 views • 3 years ago

🚨 New paper out! “FineVision: Open Data Is All You Need” 🥳 We unified 200+ data sources into 24M samples. That’s 17.3M images and 9.5B answer tokens, the largest open VLM dataset ever released. All fully documented, reproducible, and available for everyone. And there's more! 🎢

🚨 New paper out! “FineVision: Open Data Is All You Need” 🥳 We unified 200+ data sources into 24M samples. That’s 17.3M images and 9.5B answer tokens, the largest open VLM dataset ever released. All fully documented, reproducible, and available for everyone. And there's more! 🎢

Andi Marafioti

46,437 views • 7 months ago

Breaking down how the FBI is using your data for surveillance

Breaking down how the FBI is using your data for surveillance

Channel 5

41,047 views • 2 months ago