Search papers, labs, and topics across Lattice.
4
1
6
Relaxing the rigid semantic labeling of indoor spaces, allowing for "confusions" within rooms, actually *improves* a service robot's ability to find objects.
A novel self-conditioned GAN learns trajectory forecasting without context, outperforming supervised methods in poorly labeled data by discovering behavioral modes in the discriminator's feature space.
By explicitly modeling how people differ, adversarial learning can be steered to produce activity recognition models that generalize better to new users.
Ditch RANSAC for loop closure: a maximal clique search on feature correspondences delivers more robust and accurate 3D LiDAR SLAM, especially in noisy or ambiguous environments.
