SOTAVerified

Scene Segmentation

Scene segmentation is the task of splitting a scene into its various object components.

Image adapted from Temporally coherent 4D reconstruction of complex dynamic scenes.

Papers

Showing 5175 of 283 papers

TitleStatusHype
Exploring Event-driven Dynamic Context for Accident Scene SegmentationCode1
A Shared Representation for Photorealistic Driving SimulatorsCode1
Mesh Convolution with Continuous Filters for 3D Surface ParsingCode1
DSPoint: Dual-scale Point Cloud Recognition with High-frequency FusionCode1
FTNet: Feature Transverse Network for Thermal Image Semantic SegmentationCode1
Multi-Domain Incremental Learning for Semantic SegmentationCode1
CondNet: Conditional Classifier for Scene SegmentationCode1
Trans4Trans: Efficient Transformer for Transparent Object and Semantic Scene Segmentation in Real-World Navigation AssistanceCode1
Bird's-Eye-View Panoptic Segmentation Using Monocular Frontal View ImagesCode1
Standardized Max Logits: A Simple yet Effective Approach for Identifying Unexpected Road Obstacles in Urban-Scene SegmentationCode1
Unsupervised Discovery of Object Radiance FieldsCode1
Image2Point: 3D Point-Cloud Understanding with 2D Image Pretrained ModelsCode1
RobustNet: Improving Domain Generalization in Urban-Scene Segmentation via Instance Selective WhiteningCode1
Adaptive Boosting for Domain Adaptation: Towards Robust Predictions in Scene SegmentationCode1
Point TransformerCode1
SNE-RoadSeg: Incorporating Surface Normal Information into Semantic Segmentation for Accurate Freespace DetectionCode1
Scene Segmentation with Dual Relation-aware Attention NetworkCode1
Learning and Reasoning with the Graph Structure Representation in Robotic SurgeryCode1
Learning Physical Graph Representations from Visual ScenesCode1
A Local-to-Global Approach to Multi-modal Movie Scene SegmentationCode1
Context Prior for Scene SegmentationCode1
DualConvMesh-Net: Joint Geodesic and Euclidean Convolutions on 3D MeshesCode1
Fusion-Aware Point Convolution for Online Semantic 3D Scene SegmentationCode1
Cars Can't Fly up in the Sky: Improving Urban-Scene Segmentation via Height-driven Attention NetworksCode1
FPConv: Learning Local Flattening for Point ConvolutionCode1
Show:102550
← PrevPage 3 of 12Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1ICMMean IoU50.6Unverified
2Index NetworkMean IoU33.48Unverified
3DeepLab-LargeFOVMean IoU32.08Unverified
4SegNetMean IoU31.84Unverified
5FCNMean IoU27.39Unverified
#ModelMetricClaimedVerifiedStatus
13DMVAverage Accuracy75Unverified
2KPConv3DIoU68.6Unverified
3PointNet++Average Accuracy60.2Unverified
#ModelMetricClaimedVerifiedStatus
1Mask2AnomalyOpen-mIoU59.8Unverified
2LDN121-RPLOpen-mIoU56.3Unverified
3LDN121-DenseHybridOpen-mIoU45.8Unverified
#ModelMetricClaimedVerifiedStatus
1NeighborNetAP71.9Unverified
2TranS4merAP60.78Unverified
#ModelMetricClaimedVerifiedStatus
1UNetFormerCategory mIoU67.8Unverified