Towards Better Generalization: Joint Depth-Pose Learning without PoseNet
Visual Odometry Revisited: What Should Be Learnt?
Deep Virtual Stereo Odometry: Leveraging Deep Depth Prediction for Monocular Direct Sparse Odometry
D3VO: Deep Depth, Deep Pose and Deep Uncertainty for Monocular Visual Odometry
TMU实验室Deepleraning
为了解决训练图像对之间光照不一致的问题,我们的网络对训练过程中光源和目标图像的亮度变化参数进行动态预测。