Joint detection and tracking model named DEFT, or ``Detection Embeddings for Tracking." Our approach relies on an appearance-based object matching network jointly-learned with an underlying object detection network. An LSTM is also added to capture motion constraints.
I don't know why i am having this issue. All libraries and requirements are updated? Does anyone know how to fix that error?