Deep convolutional networks are prominently used in object detection tasks due to their notable performances. These typically have pooling layers following the convolution, which effectively subsamples convolution output, potentially introducing aliasing. An aliased signal emerging earlier inevitably propagates throughout network and such distortion prevents getting best performance out of a ne...

The application of RPCA model in moving object detection can accurately extract the foreground, but effect is not ideal under complex dynamic background conditions. Based on this, this paper proposes an improved based rank–1 regulation and 3D-TV. uses term to describe low rank video background, 3D-TV constrain spatiotemporal continuity objects, F-norm eliminate interference background. experime...

Mainstream object detectors are commonly constituted of two sub-tasks, including classification and regression tasks, implemented by parallel heads. This classic design paradigm inevitably leads to inconsistent spatial distributions between score localization quality (IOU). Therefore, this paper alleviates misalignment in the view knowledge distillation. First, we observe that massive teacher a...