computer-vision-1

Zero Shot Segmentation

3260 papers • 126 benchmarks • 313 datasets

This task has no description! Would you like to contribute one?

(Image credit: Papersgraph)

Benchmarks

These leaderboards are used to track progress in zero-shot-segmentation-1

Trend

Dataset

Best Model

Actions

Segmentation in the Wild

ADE20K training-free zero-shot segmentation

Libraries

i

Use these libraries to find zero-shot-segmentation-1 models and implementations

huggingface/transformers

2 papers 124,196

Datasets

Segmentation in the Wild

TomoSAM

MatSeg

Subtasks

No subtasks available.

Most implemented papers

Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection

Hang Su, Jianwei Yang, Chun-yue Li, Hao Zhang, Feng Li, Tianhe Ren, Jun-Juan Zhu, Shilong Liu, Zhaoyang Zeng, Jie Yang, Lei Zhang•Wed Mar 08 2023

An open-set object detector, called Grounding DINO, is presented by marrying Transformer-based detector DINO with grounded pre-training, which can detect arbitrary objects with human inputs such as category names or referring expressions, and performs remarkably well on all three settings.

Content

IDEA-Research/Grounded-Segment-Anyt…

2 papers 13,324

3418

0

Paper Graph

Image Segmentation Using Text and Image Prompts

Alexander S. Ecker, Timo Lüddecke•Fri Dec 17 2021

This work proposes a system that can generate image segmentations based on arbitrary prompts at test time, and builds upon the CLIP model as a backbone which it extends with a transformer-based decoder that enables dense prediction.

668 0

Paper Graph

Side Adapter Network for Open-Vocabulary Semantic Segmentation

Han Hu, Fangyun Wei, Xiang Bai, Zheng Zhang, Mengde Xu•Wed Feb 22 2023

This paper presents a new framework for open-vocabulary semantic segmentation with the pre-trained vision-language model, named Side Adapter Network (SAN), which significantly outperforms other counterparts, with up to 18 times fewer trainable parameters and 19 times faster inference speed.

371 0

Paper Graph

A Simple Framework for Open-Vocabulary Segmentation and Detection

Jianfeng Gao, Jianwei Yang, Siyi Liu, Chun-yue Li, Xueyan Zou, Hao Zhang, Feng Li, Lei Zhang•Mon Mar 13 2023

OpenSeeD is the first to explore the potential of joint training on segmentation and detection, and hope it can be received as a strong baseline for developing a single model for both tasks in the open world.

215 0

Paper Graph

Context-aware Feature Generation For Zero-shot Semantic Segmentation

Li Niu, Liqing Zhang, Siyuan Zhou, Zihan Zhao, Zhangxuan Gu•Sat Aug 15 2020

This paper proposes a novel context-aware feature generation method for zero-shot segmentation named CaGNet, which inserts a contextual module in a segmentation network to capture the pixel-wise contextual information, which guides the process of generating more diverse and context- aware features from semantic word embeddings.

154 0

Paper Graph

CLIP Surgery for Better Explainability with Enhancement in Open-Vocabulary Tasks

Yi Li, Hualiang Wang, Yiqun Duan, Xiaomeng Li•Sat Dec 31 2022

The proposed CLIP Surgery is a method that enables surgery-like modiﬁca-tions for the inference architecture and features, for better explainability and enhancement in multiple open-vocabulary tasks and demonstrates remarkable improvements in open-vocabulary segmentation and multi-label improvements.

126 0

Paper Graph

Segment Anything Model for Medical Image Analysis: an Experimental Study

N. Konz, M. Mazurowski, Haoyu Dong, Han Gu, Jichen Yang•Wed Apr 19 2023

An extensive evaluation of Segment Anything Model's ability to segment medical images on a collection of 19 medical imaging datasets from various modalities and anatomies concludes that SAM shows impressive zero-shot segmentation performance for certain medical Imaging datasets, but moderate to poor performance for others.

694 0

Paper Graph

One-Prompt to Segment All Medical Images

Junde Wu, Min Xu•Tue May 16 2023

A new paradigm toward the universal medical image segmentation, termed ‘One-Prompt Segmentation,’ which combines the strengths of one-shot and interactive methods and can adeptly handle the un-seen task in a single forward pass.

47 0

Paper Graph

Segment Anything in High Quality

F. Yu, Yu-Wing Tai, Chi-Keung Tang, Lei Ke, Martin Danelljan, Yifan Liu, Mingqiao Ye•Thu Jun 01 2023

A learnable High-Quality Output Token is injected into SAM's mask decoder and is responsible for predicting the high-quality mask, which reuses and preserves the pre-trained model weights of SAM, while only introducing minimal additional parameters and computation.

510 0

Paper Graph

Unsupervised deep learning for Bayesian brain MRI segmentation

Adrian V. Dalca, M. Sabuncu, P. Golland, Evan M. Yu, B. Fischl, J. E. Iglesias•Wed Apr 24 2019

This paper proposes an alternative strategy that combines a conventional probabilistic atlas-based segmentation with deep learning, enabling one to train a segmentation model for new MRI scans without the need for any manually segmented images.

72 0

Paper Graph

Adding a benchmark result helps the community track progress.