Загрузка видео...

Не удалось загрузить видео

На главную

Sharada manuscripts are among the most difficult Indic texts for OCR. Their lines are often long, curved, and rich in diacritics. We built CURIO, a curvature-aligned OCR to read them. 🧵👇

vikataravi's profile picture

Ravi Kiran S

1,592 subscribers

36,320 просмотров • 5 месяцев назад •via X (Twitter)

ОбразованиеНаука и технологии
LIVE

Anya Rossi• Live Now

Private livecam show

Комментарии: 0

Нет доступных комментариев

Здесь появятся комментарии из оригинального поста

Похожие видео

Inspired by Brian Roemmele, I set up DeepSeek-OCR on colab. Even with a T4 GPU and 4-bit quantization, it scans a page in about 45 seconds. In this video, you can see that it compressed 527 text tokens to 249 image tokens. DeepSeek-OCR is trained on nearly 100 languages, they say in their paper. So, I'm going to try using it on some old manuscripts written in indic languages soon.
0:44

Inspired by Brian Roemmele, I set up DeepSeek-OCR on colab. Even with a T4 GPU and 4-bit quantization, it scans a page in about 45 seconds. In this video, you can see that it compressed 527 text tokens to 249 image tokens. DeepSeek-OCR is trained on nearly 100 languages, they say in their paper. So, I'm going to try using it on some old manuscripts written in indic languages soon.

Ashraff Hathibelagal

226,767 просмотров • 9 месяцев назад

Today, we're introducing our document parser built specifically for RAG. The parser combines the best vision, OCR, and vision language models to deliver unmatched accuracy. Try it for free today—the first 500+ pages are on us! 🧵 1/
1:51

Today, we're introducing our document parser built specifically for RAG. The parser combines the best vision, OCR, and vision language models to deliver unmatched accuracy. Try it for free today—the first 500+ pages are on us! 🧵 1/

Douwe Kiela

1,308,593 просмотров • 1 год назад

We used DeepSeek OCR to extract every dataset from tables/charts across 500k+ AI arXiv papers for $1000 🚀 See which benchmarks are trending and discover datasets you didn't know existed Doing the same task with Mistral OCR would've cost $7500 👀
0:30

We used DeepSeek OCR to extract every dataset from tables/charts across 500k+ AI arXiv papers for $1000 🚀 See which benchmarks are trending and discover datasets you didn't know existed Doing the same task with Mistral OCR would've cost $7500 👀

alphaXiv

288,761 просмотров • 9 месяцев назад

Let's talk parsing charts 📊📈. Last week we released ParseBench, the first document OCR benchmark for AI agents. New in ParseBench: ChartDataPointMatch. Most document look at a chart and OCR the caption. Agents need the actual numbers. That's the gap between "OCR'd the text around the chart" and "actually read the chart." More about ParseBench, the GitHub code, Hugging Face dataset, and scientific paper→
1:21

Let's talk parsing charts 📊📈. Last week we released ParseBench, the first document OCR benchmark for AI agents. New in ParseBench: ChartDataPointMatch. Most document look at a chart and OCR the caption. Agents need the actual numbers. That's the gap between "OCR'd the text around the chart" and "actually read the chart." More about ParseBench, the GitHub code, Hugging Face dataset, and scientific paper→

LlamaIndex 🦙

13,987 просмотров • 3 месяцев назад

We raised $1.5m to launch the world’s first LLM built for developer tasks. Run OCR, web scraping, STT, and classification that you can rely on for any dev tasks. Available in Beta starting today!
1:21

We raised $1.5m to launch the world’s first LLM built for developer tasks. Run OCR, web scraping, STT, and classification that you can rely on for any dev tasks. Available in Beta starting today!

Yoeven

93,603 просмотров • 10 месяцев назад

The Democrats are running a color revolution in America. These violent acts are being promoted by their leadership to compel the most deranged among them to act.
2:15

The Democrats are running a color revolution in America. These violent acts are being promoted by their leadership to compel the most deranged among them to act.

Brenden Dilley

76,061 просмотров • 8 месяцев назад

for Basketball AI, I want to read jersey numbers and map them to player names I'm using RF-DETR + SmolVLM2 combo to do it - detect jersey numbers in real-time with RF-DETR - crop number regions - OCR the numbers with SmolVLM2 computer vision + sports:
0:21

for Basketball AI, I want to read jersey numbers and map them to player names I'm using RF-DETR + SmolVLM2 combo to do it - detect jersey numbers in real-time with RF-DETR - crop number regions - OCR the numbers with SmolVLM2 computer vision + sports:

SkalskiP

27,556 просмотров • 1 год назад

"Supermarkets in Australia are just taking the piss." Coles and Woolies are among the most profitable supermarkets in the world, and the only thing crazier than them price gouging their way through a cost of living crisis, is their excuses for it @RDNS_TAI #auspol #ausecon
1:43

"Supermarkets in Australia are just taking the piss." Coles and Woolies are among the most profitable supermarkets in the world, and the only thing crazier than them price gouging their way through a cost of living crisis, is their excuses for it @RDNS_TAI #auspol #ausecon

Australia Institute

21,595 просмотров • 2 лет назад

Toucans are among the most recognizable birds in the world, thanks to their large and colorful beaks. But it’s not just for looks, it helps them control their body temperature and reach food other birds can’t
0:28

Toucans are among the most recognizable birds in the world, thanks to their large and colorful beaks. But it’s not just for looks, it helps them control their body temperature and reach food other birds can’t

Nature Unedited

248,583 просмотров • 4 месяцев назад

Airports are crowded and security lines can stretch for what feels like miles. Some travelers have discovered a shortcut — requesting a wheelchair at check-in, which often allows them to bypass long lines and get priority assistance through the airport. You’re standing in a long line that hasn’t moved in 20 minutes. Suddenly you notice several people being wheeled past the crowd, heading straight toward the front. Wheelchair services are meant to help passengers who truly need mobility assistance, but stories of travelers using the service just to skip lines have sparked debate among frequent flyers and airport staff. If someone requests a wheelchair mainly to avoid long lines, is it a clever travel hack — or taking advantage of a service meant for those who genuinely need it? ✈️
0:59

Airports are crowded and security lines can stretch for what feels like miles. Some travelers have discovered a shortcut — requesting a wheelchair at check-in, which often allows them to bypass long lines and get priority assistance through the airport. You’re standing in a long line that hasn’t moved in 20 minutes. Suddenly you notice several people being wheeled past the crowd, heading straight toward the front. Wheelchair services are meant to help passengers who truly need mobility assistance, but stories of travelers using the service just to skip lines have sparked debate among frequent flyers and airport staff. If someone requests a wheelchair mainly to avoid long lines, is it a clever travel hack — or taking advantage of a service meant for those who genuinely need it? ✈️

TheRealCherokeeOwl 🦉

14,720 просмотров • 4 месяцев назад

The women of Keralam are the backbone of the state and an inspiration to the entire country. For millions across India and around the world, Kerala’s nurses are a source of comfort and strength in life’s most difficult moments. They have made it possible for me to campaign in Keralam with peace of mind, knowing that my mother is being cared for by them in Delhi. We are all deeply indebted to their compassion, dedication, and kindness.
1:02

The women of Keralam are the backbone of the state and an inspiration to the entire country. For millions across India and around the world, Kerala’s nurses are a source of comfort and strength in life’s most difficult moments. They have made it possible for me to campaign in Keralam with peace of mind, knowing that my mother is being cared for by them in Delhi. We are all deeply indebted to their compassion, dedication, and kindness.

Rahul Gandhi

101,045 просмотров • 4 месяцев назад

CURVED CONE COLLECT ↪️ There are a ton of different curved sprinting drills… There are a ton of different change of direction drills… ✅ This drill combines curved sprinting with COD ✅ Sports are NOT played in straight lines. Training needs to reflect that! Finding ways to bring curved sprinting and COD together in training can further develop & prepare athletes for in-game demands.
0:59

CURVED CONE COLLECT ↪️ There are a ton of different curved sprinting drills… There are a ton of different change of direction drills… ✅ This drill combines curved sprinting with COD ✅ Sports are NOT played in straight lines. Training needs to reflect that! Finding ways to bring curved sprinting and COD together in training can further develop & prepare athletes for in-game demands.

Tim Cortazzo

49,964 просмотров • 1 год назад

Chinese people are massively buying winter melons for their pets — using them to cool down cats and dogs in hot weather. The fruits are about 95% water and, thanks to their thick skin, stay cool for a long time even without refrigeration. For many, this has become a viral summer trend: pets lie on top of the melons and often start chewing on them afterward.
0:31

Chinese people are massively buying winter melons for their pets — using them to cool down cats and dogs in hot weather. The fruits are about 95% water and, thanks to their thick skin, stay cool for a long time even without refrigeration. For many, this has become a viral summer trend: pets lie on top of the melons and often start chewing on them afterward.

NEXTA

76,779 просмотров • 15 дней назад

Baidu Inc. just dropped PaddleOCR-VL-1.5, and it’s not just another OCR model. This is a fully open-source OCR model with just 0.9B parameters, built specifically for production-grade document understanding. And the results speak loudly 👇 On OmniDocBench V1.5, the most authoritative global document parsing benchmark, PaddleOCR-VL-1.5 ranks #1 overall, hitting 94.5% overall accuracy, outperforming models like DeepSeek-OCR2. As the first OCR model natively supporting irregular document layout positioning, it perfectly solves OCR pain points in production for complex structured business documents such as financial reports and insurance forms. While a lot of the industry is racing toward bigger models, PaddleOCR is betting on something smarter: small parameters + high performance + open source + real usability. That combination is exactly what scalable document intelligence needs right now. Demo / API → Open source → Model → Curious to see how teams start building on top of this.
0:29

Baidu Inc. just dropped PaddleOCR-VL-1.5, and it’s not just another OCR model. This is a fully open-source OCR model with just 0.9B parameters, built specifically for production-grade document understanding. And the results speak loudly 👇 On OmniDocBench V1.5, the most authoritative global document parsing benchmark, PaddleOCR-VL-1.5 ranks #1 overall, hitting 94.5% overall accuracy, outperforming models like DeepSeek-OCR2. As the first OCR model natively supporting irregular document layout positioning, it perfectly solves OCR pain points in production for complex structured business documents such as financial reports and insurance forms. While a lot of the industry is racing toward bigger models, PaddleOCR is betting on something smarter: small parameters + high performance + open source + real usability. That combination is exactly what scalable document intelligence needs right now. Demo / API → Open source → Model → Curious to see how teams start building on top of this.

Bishal Nandi

35,318 просмотров • 6 месяцев назад

Announcing: Agentic Document Extraction! PDF files represent information visually - via layout, charts, graphs, etc. - and are more than just text. Unlike traditional OCR and most PDF-to-text approaches, which focus on extracting the text, an agentic approach lets us break a document down into components and reason about them, resulting in more accurate extraction of the underlying meaning for RAG and other applications. Watch the video for details.
2:45

Announcing: Agentic Document Extraction! PDF files represent information visually - via layout, charts, graphs, etc. - and are more than just text. Unlike traditional OCR and most PDF-to-text approaches, which focus on extracting the text, an agentic approach lets us break a document down into components and reason about them, resulting in more accurate extraction of the underlying meaning for RAG and other applications. Watch the video for details.

Andrew Ng

689,377 просмотров • 1 год назад

🚨TRUMP: "We're going to have to override the Coastal Commission. I've dealt with the Coastal Commission for a long time and they're considered the most difficult in the country. We can't afford to wait 10 years and have them play their game. I'm going to override the Coastal Commission. I'm not going to let them get away with their antics."
0:28

🚨TRUMP: "We're going to have to override the Coastal Commission. I've dealt with the Coastal Commission for a long time and they're considered the most difficult in the country. We can't afford to wait 10 years and have them play their game. I'm going to override the Coastal Commission. I'm not going to let them get away with their antics."

Autism Capital 🧩

1,565,716 просмотров • 1 год назад

Macron warns Ukraine could fall quickly if the lines break. He isn’t wrong. This graphic shows the flow of the frontlines during World War 1 Often the lines are static for a long time until a sudden breakthrough. We need to send more weapons to Ukraine
1:32

Macron warns Ukraine could fall quickly if the lines break. He isn’t wrong. This graphic shows the flow of the frontlines during World War 1 Often the lines are static for a long time until a sudden breakthrough. We need to send more weapons to Ukraine

Visegrád 24

1,655,785 просмотров • 2 лет назад

AOC: We are allowed to have fun while we are destroying fascism. We are allowed to celebrate our victories, and in fact, oftentimes the side that is having the most fun often wins.
1:24

AOC: We are allowed to have fun while we are destroying fascism. We are allowed to celebrate our victories, and in fact, oftentimes the side that is having the most fun often wins.

Acyn

40,280 просмотров • 3 месяцев назад

Most people wait for the market. We built a way to trade before it exists. Triadmarkets is entering a new phase on Solana 🧵👇
0:39

Most people wait for the market. We built a way to trade before it exists. Triadmarkets is entering a new phase on Solana 🧵👇

Triadmarkets

105,648 просмотров • 5 месяцев назад