Details for method 'DCNAS+ASPP [Mapillary Vistas]'
|name||DCNAS+ASPP [Mapillary Vistas]|
|challenge||pixel-level semantic labeling|
|details||Existing NAS algorithms usually compromise on restricted search space or search on proxy task to meet the achievable computational demands. To allow as wide as possible network architectures and avoid the gap between realistic and proxy setting, we propose a novel Densely Connected NAS (DCNAS) framework, which directly searches the optimal network structures for the multi-scale representations of visual information, over a large-scale target dataset without proxy. Specifically, by connecting cells with each other using learnable weights, we introduce a densely connected search space to cover an abundance of mainstream network designs. Moreover, by combining both path-level and channel-level sampling strategies, we design a fusion module and mixture layer to reduce the memory consumption of ample search space, hence favor the proxyless searching. Compared with contemporary works, experiments reveal that the proxyless searching scheme is capable of bridge the gap between searching and training environments.|
|project page / code|
|used Cityscapes data||fine annotations, coarse annotations|
|used external data||mapillary vistas|
|submission date||November, 2020|