Convolutional Neural Networks

September 17, 2025 ยท View on GitHub

ModelDescriptionVenueYear
StereoSRparallax prior, jointly training two-stage networks for luminance and chrominanceCVPR2018
PASSRnetparallax-attention mechanism with a global receptive field along the epipolar lineCVPR2019
SAMgeneric stereo attention, arbitrary SRSPL2020
DCSSRatrous parallax-attentionarXiv2020
DASSRDisparity-Aware Domain AdaptationCVPR2020
FMDBlearned the inherent pixel correspondenceCVPR2020
SPAMstereo-consistency constraintAAAI2020
IMinteraction moduleTCSVT2020
iPASSRsymmetric bi-directional parallax attention moduleCVPR2021
Cross-MPIMultiplane ImagesCVPR2021
CPASSRnetsingle model performs multiple scale factors for both viewsTMM2021
CVCnetglobal contextual and local features extracted from both viewsTMM2021
BSSRnetbilateral grid processingSPL2021
DFAMdisparity feature alignment moduleSPL2021
SSRDE-FNetsimultaneously handles the stereo image super-resolution and disparity estimationACMMM2021
NAFSSRbased on NAFNet, cross attention modules to fuse featuresCVPRW2022

Transformer-based Methods

ModelDescriptionPublicationYear
Trans-SVSRself-attention and optical flow-based, TransformerCVPRW2022
SwiniPASSRSwin TransformerCVPRW2022
PFT-SSRSwin TransformerICASSP2023