Hugging Face's logo

Kornia AI
non-profit
http://kornia.org
kornia_foss
kornia
Activity Feed Request to join this org

AI & ML interests

computer-vision, image-processing, machine-learning, deep-learning

Recent Activity

gagan3012  authored a paper 9 days ago
Leveraging Vision-Language Pre-training for Human Activity Recognition in Still Images
marianna13  authored a paper 19 days ago
Alice in Wonderland: Simple Tasks Showing Complete Reasoning Breakdown in State-Of-the-Art Large Language Models
marianna13  authored a paper 19 days ago
DataComp-LM: In search of the next generation of training sets for language models
View all activity

merve 
posted an update about 6 hours ago
view post
Post
136
Dataset Viewer for PDFs just landed on Hugging Face 📖🤗 you can now preview all the PDFs easier than before!

on top of this, there's PdfFolder format to load the PDF datasets quicker 💨
> to use it, your dataset should follow a directory format like folder/train/doc1.pdf, folder/train/doc1.pdf
> if you want to include bounding boxes, labels etc. you can keep them in a metadata.csv file in the same folder 🤝

read document dataset docs https://huggingface.co/docs/datasets/main/en/document_dataset
check all the document datasets here https://huggingface.co/datasets?modality=modality:document&sort=trending 📖
prithivMLmods 
posted an update 2 days ago
view post
Post
1424
The demo for DREX-062225-exp (Document Retrieval and Extraction eXpert ~ experimental) / typhoon-ocr-3b (a bilingual document parsing model built specifically for real-world documents) / VIREX-062225-exp (Video Information Retrieval and Extraction eXpert ~ experimental) / olmOCR-7B-0225-preview (the document parsing model based on Qwen2VL). 🤗

✦ Demo : prithivMLmods/Doc-VLMs-OCR ~ ( with .md canvas )

⤷ DREX-062225-exp : prithivMLmods/DREX-062225-exp
⤷ typhoon-ocr-3b : scb10x/typhoon-ocr-3b
⤷ VIREX-062225-exp : prithivMLmods/VIREX-062225-exp
⤷ olmOCR-7B-0225-preview : allenai/olmOCR-7B-0225-preview

⤷ Collection : prithivMLmods/doc-vl-685839064a863e1cd23be3f1
⤷ Multimodal Implementations : prithivMLmods/multimodal-implementations-67c9982ea04b39f0608badb0
.
.
.

To know more about it, visit the model card of the respective model. !!
  • 2 replies
·
merve 
posted an update 2 days ago
view post
Post
461
we've merged LightGlue keypoint matcher to Hugging Face transformers! it allows commercial use when paired with an open-source keypoint detector 🙏🏻

it works very well, try it yourself: ETH-CVG/LightGlue

here's an in-the-wild test with two images of the same place ⤵️
  • 1 reply
·
prithivMLmods 
posted an update 3 days ago
view post
Post
2574
Updated the docscopeOCR-7B-050425-exp with the DREX-062225-exp, with improved preciseness in table structure and line spacing in the markdown used on the document page. And though this is still an experimental one, it's expected to perform well in the defined DREX use cases [ Document Retrieval and Extraction eXpert – experimental ocr ]. 💻

⤷ Model : prithivMLmods/DREX-062225-exp
⤷ Demo : prithivMLmods/Doc-VLMs-OCR

⤷ Collection : prithivMLmods/doc-vl-685839064a863e1cd23be3f1
⤷ Multimodal Implementations : prithivMLmods/multimodal-implementations-67c9982ea04b39f0608badb0
⤷ Git : https://github.com/PRITHIVSAKTHIUR/DREX.git
.
.
.

To know more about it, visit the model card of the respective model. !!
merve 
posted an update 3 days ago
view post
Post
4185
Release picks of the past week is here! Find more models, datasets, Spaces here merve/june-20-releases-68594824d1f4dfa61aee3433

🖼️ VLMs/OCR
> moonshotai/Kimi-VL-A3B-Thinking-2506 is a powerful reasoning vision LM, 3B active params, smarter with less tokens, supports long documents, videos 👏 (OS)
> nanonets/Nanonets-OCR-s is 3.75B params OCR model based on Qwen2.5VL-3B-Instruct (OS)

💬 LLMs
> moonshotai/Kimi-Dev-72B is a strong coding model based on Qwen2.5-72B (OS)
> Mistral released mistralai/Mistral-Small-3.2-24B-Instruct-2506, an update to their former model with better function calling & instruction following (OS)

🗣️ Audio
> Google released google/magenta-realtime, real time music generation & audio synthesis (cc-by-4)
> kyutai released new speech-to-text models that come in 1B & 2B ( kyutai/stt-1b-en_fr, stt-2b-en_fr) with 0.5s and 2.5s delay

3D
> Tencent released tencent/Hunyuan3D-2.1 an image-to-3D model (see below)
merve 
posted an update 4 days ago
merve 
posted an update 6 days ago
prithivMLmods 
posted an update 6 days ago
view post
Post
1749
The demo for smoldocling / nanonets ocr / typhoon ocr / monkey ocr explores the document OCR capabilities of various newly released multimodal VLMs in a single space. And if you're experiencing or demoing long document image OCR, kindly use the Smoldocling 256M preview [ Smoldocling is back in demo here. ] 🤗.

✦ Try the demo here : prithivMLmods/Multimodal-OCR2

⤷ MonkeyOCR Recognition : echo840/MonkeyOCR
⤷ Nanonets-OCR-s : nanonets/Nanonets-OCR-s
⤷ SmolDocling-256M-preview : ds4sd/SmolDocling-256M-preview
⤷ typhoon-ocr-7b : scb10x/typhoon-ocr-7b

⤷ Multimodal Implementations : prithivMLmods/multimodal-implementations-67c9982ea04b39f0608badb0

⤷ Github : https://github.com/PRITHIVSAKTHIUR/Multimodal-OCR2


The community GPU grant was given by Hugging Face — special thanks to them. 🤗🚀



To know more about it, visit the model card of the respective model. !!
  • 2 replies
·
merve 
posted an update 7 days ago
view post
Post
1849
stop using VLMs blindly ✋🏻

compare different VLM outputs on a huge variety of inputs (from reasoning to OCR!) 🔥 visionLMsftw/comparevlms

> has support for multiple VLMs: google/gemma-3-27b-it, Qwen/Qwen2.5-VL-7B-Instruct, Qwen/Qwen2.5-VL-32B-Instruct, meta-llama/Llama-4-Maverick-17B-128E-Instruct, HuggingFaceTB/SmolVLM2-2.2B-Instruct
> recommend us new models or inputs, we'll add 🫡

so far I figured out
> for fact-checks, you need a relatively bigger size (7B is ok!)
> Gemma 3 gets downgrade without pan and scan (especially for 📑)
> Qwen2.5VL-32B is very talkative, great for reasoning but not good for simple tasks 🗣️
  • 2 replies
·
merve 
posted an update 8 days ago
view post
Post
3566
Releases of the past week are here merve/releases-june-13-6852c3c1eaf1e0c24c958860

Here's our picks 🤓
So many interesting models released past week in open AI! 🤖

🖼️ Computer Vision/VLMs
> nanonets/Nanonets-OCR-s is the new state-of-the-art OCR model that can handle checkboxes, watermarks, tables (OS)
> Meta released facebook/v-jepa-2-6841bad8413014e185b497a6, new sota video embeddings with two new classification models (OS)
> ByteDance-Seed/SeedVR2-3B is a new 3B video restoration model (OS)

Audio
> Stepfun released stepfun-ai/Step-Audio-AQAA, new large (137B 🤯) audio language model that takes in audio and generates audio (OS)

🤖 Robotics
> nvidia released nvidia/GR00T-N1.5-3B, new open foundation vision language action model

3D
> tencent/Hunyuan3D-2.1 is the new version of Hunyuan by Tencent that can generate 3D assets from text and image prompts
gagan3012 
authored a paper 9 days ago
prithivMLmods 
posted an update 9 days ago
view post
Post
3728
The demo for the MonkeyOCR Recognition model, which adopts a Structure-Recognition-Relation (SRR) triplet paradigm & Nanonets-OCR-s a powerful, state-of-the-art image-to-markdown OCR model that goes far beyond traditional text extraction and other experimental document OCR models, is combined into a single space.

✦ Try the demo here : prithivMLmods/core-OCR
✦ Try Nanonets-OCR-s demo here : prithivMLmods/Multimodal-OCR

⤷ MonkeyOCR Recognition : echo840/MonkeyOCR
⤷ docscopeOCR-7B-050425-exp : prithivMLmods/docscopeOCR-7B-050425-exp
⤷ coreOCR-7B-050325-preview : prithivMLmods/coreOCR-7B-050325-preview
⤷ Nanonets-OCR-s : nanonets/Nanonets-OCR-s

⤷ Multimodal Implementations : prithivMLmods/multimodal-implementations-67c9982ea04b39f0608badb0

Also, include a sample OCR test using the VisionOCR-3B-061125 model and the Qwen2-VL-OCR-2B-Instruct model.
⤷ Blog : https://huggingface.co/blog/prithivMLmods/visionocr-3b-061125-vs-qwen2-vl-ocr-2b-instruct

To know more about it, visit the model card of the respective model. !!
merve 
posted an update 9 days ago
view post
Post
3502
IN: video fine-tuning support for facebook V-JEPA 2 in HF transformers 🔥

it comes with
> four models fine-tuned on Diving48 and SSv2 dataset facebook/v-jepa-2-6841bad8413014e185b497a6
> FastRTC demo on V-JEPA2 SSv2 qubvel-hf/vjepa2-streaming-video-classification
> fine-tuning script on UCF-101 https://gist.github.com/ariG23498/28bccc737c11d1692f6d0ad2a0d7cddb
> fine-tuning notebook on UCF-101 https://colab.research.google.com/drive/16NWUReXTJBRhsN3umqznX4yoZt2I7VGc?usp=sharing
we're looking forward to see what you will build! 🤗
merve 
posted an update 10 days ago
view post
Post
2418
#CVPR2025 Paper Picks #1
VisionZip is a compression technique that reduces number of visual tokens to improve performance AND prefill time for vision language models
demo: Senqiao/VisionZip
paper: VisionZip: Longer is Better but Not Necessary in Vision Language Models (2412.04467)
most of the image tokens are redundant for the LLM, so the authors ask "are all visual tokens necessary?"

the method is simple:
find which tokens have the highest attention score, merge rest of the tokens based on similarity, then merge both

their method is both training-free and for fine-tuning
the authors report 5 point improvement on average of vision language tasks + 8x improvement in prefilling time for Llava-Next 7B and 13B 🤯

removing redundant tokens improve image token quality too 🥹
merve 
posted an update 10 days ago
view post
Post
3639
stop writing CUDA kernels yourself

we have launched Kernel Hub: easy optimized kernels for all models on Hugging Face 🔥 use them right away!
it's where the community populates optimized kernels 🤝

this release comes in three parts
> Kernel Hub: contains (as of now) 14 kernels
> kernels: Python library to load kernels from Kernel Hub
> kernel-builder: Nix package to build kernels for PyTorch (made using PyTorch C++ frontend)

when building models, your regular workflow should be pulling kernels from Hub and building your model with them 🤗
here's a practical example with RMSNorm:
1. pull the kernel from Hub with get_kernel
2. decorate with use_kernel_forward_from_hub
3. inject it to your model
we'd love to hear your feedback! 🙏🏻
we also welcome kernel contributions by community 🥹💗

- request kernels here: kernels-community/README#1
- check out this org: kernels-community
- read the blog: https://huggingface.co/blog/hello-hf-kernels
  • 1 reply
·
merve 
posted an update 13 days ago
view post
Post
697
Dolphin: new OCR model by ByteDance with MIT license 🐬

the model first detects element in the layout (table, formula etc) and then parses each element in parallel for generation
Model: ByteDance/Dolphin
Try the demo: ByteDance/Dolphin
merve 
posted an update 15 days ago
view post
Post
1364
stop building parser pipelines 👋🏻
there's a new document parser that is small, fast, Apache 2.0 licensed and is better than all the other ones! 😱

echo840/MonkeyOCR is a 3B model that can parse everything (charts, formules, tables etc) in a document 🤠
> the authors show in the paper that document parsing pipelines often have errors propagating back
> using singular e2e models are better but they're too heavy to use

this model addresses both: it's lighter, faster, stronger 🔥
merve 
posted an update 15 days ago
view post
Post
1595
Meta just released V-JEPA 2: new open-source image/video world models ⏯️🤗 facebook/v-jepa-2-6841bad8413014e185b497a6

> based on ViT, different sizes (L/G/H) and resolution (286/384)
> 0-day support in 🤗 transformers
> comes with a physical reasoning (from video) benchmark: MVPBench, IntPhys 2, and CausalVQA facebook/physical_reasoning_leaderboard

Read more https://ai.meta.com/blog/v-jepa-2-world-model-benchmarks/
We will release a fine-tuning notebook with task-specific models in transformers format soon, stay tuned!
marianna13 
authored 2 papers 19 days ago