Video yükleniyor...

Video Yüklenemedi

Ana Sayfaya Dön

Occlusion is still a challenge for SAM2. Video object segmentation is not solved yet

YuXiang_IRVL's profile picture

Yu Xiang

5,803 subscribers

53,676 görüntüleme • 1 yıl önce •via X (Twitter)

Bilim & TeknolojiEğitim
LIVE

Anya Rossi• Live Now

Private livecam show

0 Yorum

Yorum bulunmuyor

Orijinal gönderinin yorumları burada görünecek

Benzer Videolar

Today we're releasing the Segment Anything Model (SAM) — a step toward the first foundation model for image segmentation. SAM is capable of one-click segmentation of any object from any photo or video + zero-shot transfer to other segmentation tasks ➡️
0:15

Today we're releasing the Segment Anything Model (SAM) — a step toward the first foundation model for image segmentation. SAM is capable of one-click segmentation of any object from any photo or video + zero-shot transfer to other segmentation tasks ➡️

AI at Meta

3,570,334 görüntüleme • 3 yıl önce

🚀 The Segment Anything Model (SAM) has been upgraded to SAM2, featuring an efficient image encoder for segmenting images and videos. But does SAM2 outperform SAM1 in medical image and video segmentation? We're thrilled to present our paper "Segment Anything in Medical Images and Videos: Benchmark and Deployment"! We comprehensively benchmark SAM2 across 11 medical image modalities and videos. 📄 Paper: 💻 Code: **Highlights:** 1. SAM2 doesn’t always outperform SAM1 in 2D medical images, but excels in video segmentation, making it more accurate and efficient for 3D images, such as CT and MR scans. 2. MedSAM still outperforms SAM2 on most 2D modalities, but SAM2 surpasses MedSAM for 3D image segmentation in a slice-by-slice approach. 3. Segmentation performance varies with model size; sometimes the smallest model outperforms larger ones. 4. Fine-tuning SAM2 significantly boosts its performance for medical image segmentation. While SAM2 may struggle with challenging objects that have unclear boundaries or low contrast, it excels in generating good initial segmentation masks for common medical images and videos. However, the official interface doesn’t support medical data formats and has limitations on video length. To address this, we've developed a 3D Slicer Plugin and Gradio API for efficient 3D medical image and video segmentation. We invite you to try them out and provide feedback! 🔧 Deployment: - 3D Slicer Plugin: - Gradio API: (Note: Due to GPU limitations, the online API is available for only 12 hours and may be slow. We highly recommend deploying the Gradio API with your own computing resources: A big shoutout to Jun Ma (JunMa) who recently joined our UHN AI hub (UHN AI Hub) as Machine Learning Lead, and kudos to all co-authors: Sumin Kim, Feifei Li, Mohammed Baharoon (Mohammed Baharoon), Reza Asakereh, and Hongwei Lyu! This is true teamwork! Looking forward to collaborating with the community to advance 3D medical image and video segmentation foundation models! University Health Network U of T Department of Computer Science Department of Laboratory Medicine & Pathobiology Temerty Centre for AI in Medicine (T-CAIREM) Vector Institute #MedTech #AIinHealthcare #DeepLearning #MedicalImaging #SAM2 #MedSAM #AIResearch
0:22

🚀 The Segment Anything Model (SAM) has been upgraded to SAM2, featuring an efficient image encoder for segmenting images and videos. But does SAM2 outperform SAM1 in medical image and video segmentation? We're thrilled to present our paper "Segment Anything in Medical Images and Videos: Benchmark and Deployment"! We comprehensively benchmark SAM2 across 11 medical image modalities and videos. 📄 Paper: 💻 Code: **Highlights:** 1. SAM2 doesn’t always outperform SAM1 in 2D medical images, but excels in video segmentation, making it more accurate and efficient for 3D images, such as CT and MR scans. 2. MedSAM still outperforms SAM2 on most 2D modalities, but SAM2 surpasses MedSAM for 3D image segmentation in a slice-by-slice approach. 3. Segmentation performance varies with model size; sometimes the smallest model outperforms larger ones. 4. Fine-tuning SAM2 significantly boosts its performance for medical image segmentation. While SAM2 may struggle with challenging objects that have unclear boundaries or low contrast, it excels in generating good initial segmentation masks for common medical images and videos. However, the official interface doesn’t support medical data formats and has limitations on video length. To address this, we've developed a 3D Slicer Plugin and Gradio API for efficient 3D medical image and video segmentation. We invite you to try them out and provide feedback! 🔧 Deployment: - 3D Slicer Plugin: - Gradio API: (Note: Due to GPU limitations, the online API is available for only 12 hours and may be slow. We highly recommend deploying the Gradio API with your own computing resources: A big shoutout to Jun Ma (JunMa) who recently joined our UHN AI hub (UHN AI Hub) as Machine Learning Lead, and kudos to all co-authors: Sumin Kim, Feifei Li, Mohammed Baharoon (Mohammed Baharoon), Reza Asakereh, and Hongwei Lyu! This is true teamwork! Looking forward to collaborating with the community to advance 3D medical image and video segmentation foundation models! University Health Network U of T Department of Computer Science Department of Laboratory Medicine & Pathobiology Temerty Centre for AI in Medicine (T-CAIREM) Vector Institute #MedTech #AIinHealthcare #DeepLearning #MedicalImaging #SAM2 #MedSAM #AIResearch

Bo Wang

178,455 görüntüleme • 1 yıl önce

I feel strongly about this. I mean, not for git per se, but for many things. A lot of tech is "solved" on the technical / lower / system level, but not yet solved well on the higher / application level. We need many Steve Jobs for so many things!
0:12

I feel strongly about this. I mean, not for git per se, but for many things. A lot of tech is "solved" on the technical / lower / system level, but not yet solved well on the higher / application level. We need many Steve Jobs for so many things!

Hasen Judi

19,178 görüntüleme • 1 yıl önce

🚀 Wan 2.6 is now live on fal ! • Text-to-Video & Image-to-Video up to 1080p • Up to 15 second generations • Multi-shot video with intelligent scene segmentation • Import your own audio • Reference-to-Video - use 1-3 reference videos for character/object consistency
0:15

🚀 Wan 2.6 is now live on fal ! • Text-to-Video & Image-to-Video up to 1080p • Up to 15 second generations • Multi-shot video with intelligent scene segmentation • Import your own audio • Reference-to-Video - use 1-3 reference videos for character/object consistency

fal

64,542 görüntüleme • 6 ay önce

A slight increase… yet the space for comfort is still unfinished. #LittleNightmares III
0:12

A slight increase… yet the space for comfort is still unfinished. #LittleNightmares III

Little Nightmares

38,355 görüntüleme • 5 ay önce

This challenge is not for everyone 😄
0:18

This challenge is not for everyone 😄

Instablog9ja

15,793,821 görüntüleme • 2 yıl önce

🆕🎥 IT’S NOT OVER 🎉 “This summer, fate is bringing a new challenge before them…” “#MerTeş love story… is not over yet” ✨ AND EYLÜL IS COMING 👀 SEASON 2 IS OFFICIALLY ON THE WAY 🎉🔥 ARE WE READY FOR WHAT’S NEXT?! #EminGünenç #Arafta
0:17

🆕🎥 IT’S NOT OVER 🎉 “This summer, fate is bringing a new challenge before them…” “#MerTeş love story… is not over yet” ✨ AND EYLÜL IS COMING 👀 SEASON 2 IS OFFICIALLY ON THE WAY 🎉🔥 ARE WE READY FOR WHAT’S NEXT?! #EminGünenç #Arafta

Emin Günenç Global

22,863 görüntüleme • 2 ay önce

Accuracy challenge and not a cover for why my axe is red
0:16

Accuracy challenge and not a cover for why my axe is red

Gia Armani

249,816 görüntüleme • 1 yıl önce

Segment Anything is fun and powerful but still requires full mask annotation for supervised training. Our new work MaskFreeVIS appearing at #CVPR2023 shows we can get accurate video instance segmentation WITHOUT any video or image mask annotations. More at
0:20

Segment Anything is fun and powerful but still requires full mask annotation for supervised training. Our new work MaskFreeVIS appearing at #CVPR2023 shows we can get accurate video instance segmentation WITHOUT any video or image mask annotations. More at

Fisher Yu

41,103 görüntüleme • 3 yıl önce

My game is for HIDDEN OBJECT FANS, algorithm PLEASE FIND THEM! My game is for HIDDEN OBJECT FANS, algorithm PLEASE FIND THEM! My game is for HIDDEN OBJECT FANS, algorithm PLEASE FIND THEM! My game is for HIDDEN OBJECT FANS, algorithm PLEASE FIND THEM!
0:15

My game is for HIDDEN OBJECT FANS, algorithm PLEASE FIND THEM! My game is for HIDDEN OBJECT FANS, algorithm PLEASE FIND THEM! My game is for HIDDEN OBJECT FANS, algorithm PLEASE FIND THEM! My game is for HIDDEN OBJECT FANS, algorithm PLEASE FIND THEM!

Lost and Found Co. 🐤

47,996 görüntüleme • 6 ay önce

This video is so funny and yet still thinspo for me and the comments under it even more so
0:11

This video is so funny and yet still thinspo for me and the comments under it even more so

bri✌🏽👶🏽

52,231 görüntüleme • 2 ay önce

tony stark has a billion dollar flying suit but still dropped to his knees for a noise complaint. this is LITERALLY the funniest ai video yet.
0:15

tony stark has a billion dollar flying suit but still dropped to his knees for a noise complaint. this is LITERALLY the funniest ai video yet.

ViralOps

1,640,015 görüntüleme • 4 ay önce

Check out the tumble weed action on this New Mexico land spout! A land spout is a #tornado not associated with a mesocyclone occlusion
0:13

Check out the tumble weed action on this New Mexico land spout! A land spout is a #tornado not associated with a mesocyclone occlusion

Reed Timmer, PhD

59,080 görüntüleme • 1 yıl önce

WIWY challenge is not easy for everyone 😭😭
0:16

WIWY challenge is not easy for everyone 😭😭

ra ra girl³²⁷

24,320 görüntüleme • 1 yıl önce

We are presenting **SAD** (Segment Any RGBD): SAD is able to perform 3D segmentation (segment out any 3D object) with RGBD inputs (or rendered depth images only). - Code: - Demo Hugging Face:
0:11

We are presenting **SAD** (Segment Any RGBD): SAD is able to perform 3D segmentation (segment out any 3D object) with RGBD inputs (or rendered depth images only). - Code: - Demo Hugging Face:

Ziwei Liu

195,642 görüntüleme • 3 yıl önce

Posted this in insta, tw is not great for video posts but still🧌
0:11

Posted this in insta, tw is not great for video posts but still🧌

𝑺𝖆𝖗𝖆𝖍 ⚜️🏰

37,033 görüntüleme • 6 ay önce

That Sheebah challenge is still on
0:13

That Sheebah challenge is still on

BigEyeUG

24,295 görüntüleme • 2 ay önce

This challenge is not for every couple👀🤭🥱
0:19

This challenge is not for every couple👀🤭🥱

Zanele Mokoena 🇿🇦

47,594 görüntüleme • 1 ay önce

This challenge is not for the single sha 😂😂
0:15

This challenge is not for the single sha 😂😂

BIG LUCKY 𝕏

19,688 görüntüleme • 2 yıl önce