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.
|TREND||DATASET||BEST METHOD||PAPER TITLE||PAPER||CODE||COMPARE|
Online semantic 3D segmentation in company with real-time RGB-D reconstruction poses special challenges such as how to perform 3D convolution directly over the progressively fused 3D geometric data, and how to smartly fuse information from frame to frame.
This paper exploits the intrinsic features of urban-scene images and proposes a general add-on module, called height-driven attention networks (HANet), for improving semantic segmentation for urban-scene images.
#6 best model for Semantic Segmentation on Cityscapes test (using extra training data)
Specifically, we append two types of attention modules on top of traditional dilated FCN, which model the semantic interdependencies in spatial and channel dimensions respectively.
#2 best model for Semantic Segmentation on COCO-Stuff test
We present 3DMV, a novel method for 3D semantic scene segmentation of RGB-D scans in indoor environments using a joint 3D-multi-view prediction network.
SOTA for Scene Segmentation on ScanNet
Experimental results on various 3D scenes show the effectiveness of our method on 3D instance segmentation, and we also evaluate the capability of SGPN to improve 3D object detection and semantic segmentation results.
SOTA for 3D Object Detection on NYU Depth v2
With the introduction of Fully Convolutional Neural Network, which uses finer strides and utilizes deconvolutional layers for upsampling, it has been a go to for any image segmentation task.