Lin Lianpeng

Question about modulated deform convolution using

2

In paper "we ignore the modulation \_delta m_k " .Is it meaning that use deform conv as dcn_v1? In code ,i find the mask is used in ModulatedDeformConv. thx~

i find that the deform conv parameter “group” is different between "train_EDVR_woTSA_M.yml"and "train_EDVR_M.yml".why? what is the motivation? thx~

num_gpus_per_job: 1 # number of gpus each job need num_cpus_per_job: 4 # number of gpus each job need num_hosts_per_job: 1 memory_per_job: 32 # number of gpus each job need gpu_type:...