Countdown to #ICLR2020. Today's blog lets you take a visual tour of the conference portal to see what's the virtual format looks like and how to navigate our first virtual machine learning conference. Meet you at a poster or social soon 🎉 medium.com/@iclr_conf/cou…

CNN Explainer is an interactive visualization tool for learning purposes. It runs a pre-tained CNN in the browser and lets you explore the layers and operations: poloclub.github.io/cnn-explainer/ Video: youtube.com/watch?v=udVN7f… Code: github.com/poloclub/cnn-e… Paper: arxiv.org/abs/2004.15004

Multi-Scale Vision Longformer: A New Vision Transformer for High-Resolution Image Encoding pdf: arxiv.org/pdf/2103.15358… abs: arxiv.org/abs/2103.15358

CrossViT: Cross-Attention Multi-Scale Vision Transformer for Image Classification pdf: arxiv.org/pdf/2103.14899… abs: arxiv.org/abs/2103.14899

Rethinking Spatial Dimensions of Vision Transformers Proposes PiT, a self-attention model whose spatial dimension progressively shrinks like CNN, outperforms ViT. abs: arxiv.org/abs/2103.16302 code: github.com/naver-ai/pit

Check out our new transformer-based vision recognition backbone: #PiT.

Happy to share our “Rainbow Memory“ to propose a new method and evaluation protocol for more realistic CIL, #blurryCIL, which will appear @cvpr2021. Congrats to Jihwan Bang, Heesu Kim, Youngjoon Yoo, Jonghyun Choi arxiv.org/abs/2103.17230 github.com/clovaai/rainbo…

Sharing ideas on how to disseminate your research. "I am THRILLED to share that our paper is accepted to ..." Congrats! So what's next? No one is going to browse through the list of thousands of accepted papers. Ain't nobody got time for that. Check out 🧵below for examples.

Our mobile line segment detector (M-LSD) seems to generalize to anime and sketch figures pretty well! Try it out with Gradio web demo :) Paper: arxiv.org/abs/2106.00186 Github: github.com/navervision/ml… Gradio demo: gradio.app/g/AK391/mlsd (thanks to AK moved to @_akhaliq )

Happy to announce that our paper got accepted in #ICCV2021 ! We are preparing to release the code. - Learning with Memory-based Virtual Classes for Deep Metric Learning arxiv.org/abs/2103.16940

Line segment detection with Mobile LSD (M-LSD) now on Hugging Face Spaces with Gradio join beta for spaces: huggingface.co/spaces/akhaliq…

Happy to announce that our paper "Towards Light-weight and Real-time Line Segment Detection" has been accepted to #AAAI2022! Try it out in the below links :) Paper: arxiv.org/abs/2106.00186 Github: github.com/navervision/ml… Gradio demo: gradio.app/g/AK391/mlsd

#cvpr2022 accepted paper id list drive.google.com/file/d/15JFhfP… Congrats to all accepted paper authors!

Happy to share our KELIP -- Korean and English bilingual multimodal model. KELIP is trained with 1.1B image-text pairs, which are three times larger than CLIP. Try out pre-trained KELIP! paper: arxiv.org/abs/2203.14463 github: github.com/navervision/KE… demo: huggingface.co/spaces/navervi…