Figure 3 An example of a multi-stage...

Figure 3

$An example of a multi-stage context modeling scheme with three groups:{l}, {3,6,8}, {2,4,5,7}, aimed at enhancing the accuracy of probability estimation in an entropy model by exploiting spatial correlations among groups. 0 symbolizes the context with the same shape as the input point cloud, where all attribute values are 0. Yth represents the hyperprior context output by decoder hs, and Z^g defines the global hyperprior. The entropy decoding decodes the bitstream into voxels using the entropy model parameters{μ, θ} generated by gep and ɡcm. Refer to the image caption for details.$

An example of a multi-stage context modeling scheme with three groups:{l}, {3,6,8}, {2,4,5,7}, aimed at enhancing the accuracy of probability estimation in an entropy model by exploiting spatial correlations among groups. 0 symbolizes the context with the same shape as the input point cloud, where all attribute values are 0. Y_th represents the hyperprior context output by decoder h_s, and ${\hat{Z}}_{g}$ defines the global hyperprior. The entropy decoding decodes the bitstream into voxels using the entropy model parameters{μ, θ} generated by g_ep and ɡ_cm.