SOTAVerified

Instance Segmentation

Instance Segmentation is a computer vision task that involves identifying and separating individual objects within an image, including detecting the boundaries of each object and assigning a unique label to each object. The goal of instance segmentation is to produce a pixel-wise segmentation map of the image, where each pixel is assigned to a specific object instance.

Image Credit: Deep Occlusion-Aware Instance Segmentation with Overlapping BiLayers, CVPR'21

Papers

Showing 101125 of 2262 papers

TitleStatusHype
Box-supervised Instance Segmentation with Level Set EvolutionCode2
Mask2Former for Video Instance SegmentationCode2
DiffusionInst: Diffusion Model for Instance SegmentationCode2
Diving into Underwater: Segment Anything Model Guided Underwater Salient Instance Segmentation and A Large-scale DatasetCode2
CAS-ViT: Convolutional Additive Self-attention Vision Transformers for Efficient Mobile ApplicationsCode2
MaskTerial: A Foundation Model for Automated 2D Material Flake DetectionCode2
A Simple Latent Diffusion Approach for Panoptic Segmentation and Mask InpaintingCode2
DenseNets Reloaded: Paradigm Shift Beyond ResNets and ViTsCode2
Delineate Anything: Resolution-Agnostic Field Boundary Delineation on Satellite ImageryCode2
Aerial Lifting: Neural Urban Semantic and Building Instance Lifting from Aerial ImageryCode2
DetectoRS: Detecting Objects with Recursive Feature Pyramid and Switchable Atrous ConvolutionCode2
Does Image Anonymization Impact Computer Vision Training?Code2
DatasetDM: Synthesizing Data with Perception Annotations Using Diffusion ModelsCode2
DAMamba: Vision State Space Model with Dynamic Adaptive ScanCode2
Panoptic NeRF: 3D-to-2D Label Transfer for Panoptic Urban Scene SegmentationCode2
PartGLEE: A Foundation Model for Recognizing and Parsing Any ObjectsCode2
DAT++: Spatially Dynamic Vision Transformer with Deformable AttentionCode2
PTQ4SAM: Post-Training Quantization for Segment AnythingCode2
Rethinking End-to-End 2D to 3D Scene Segmentation in Gaussian SplattingCode2
Rethinking Patch Dependence for Masked AutoencodersCode2
Crowd-SAM: SAM as a Smart Annotator for Object Detection in Crowded ScenesCode2
RSPrompter: Learning to Prompt for Remote Sensing Instance Segmentation based on Visual Foundation ModelCode2
SAMRS: Scaling-up Remote Sensing Segmentation Dataset with Segment Anything ModelCode2
Scene-Centric Unsupervised Panoptic SegmentationCode2
DaViT: Dual Attention Vision TransformersCode2
Show:102550
← PrevPage 5 of 91Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1InternImage-HAP5080.8Unverified
2ResNeSt-200 (multi-scale)AP5070.2Unverified
3CenterMask + VoVNetV2-99 (multi-scale)AP5066.2Unverified
4CenterMask + VoVNetV2-57 (single-scale)AP5060.8Unverified
5Co-DETRmask AP57.1Unverified
6CBNetV2 (EVA02, single-scale)mask AP56.1Unverified
7ISDA (ResNet-50)APL55.7Unverified
8EVAmask AP55.5Unverified
9FD-SwinV2-Gmask AP55.4Unverified
10Mask Frozen-DETRmask AP55.3Unverified
#ModelMetricClaimedVerifiedStatus
1InternImage-BGFLOPs501Unverified
2Co-DETRmask AP56.6Unverified
3ViT-CoMer-L (Mask RCNN, DINOv2)mask AP55.9Unverified
4InternImage-Hmask AP55.4Unverified
5EVAmask AP55Unverified
6Mask Frozen-DETRmask AP54.9Unverified
7MasK DINO (SwinL, multi-scale)mask AP54.5Unverified
8ViT-Adapter-L (HTC++, BEiTv2, O365, multi-scale)mask AP54.2Unverified
9GLEE-Promask AP54.2Unverified
10SwinV2-G (HTC++)mask AP53.7Unverified