Search Results for author:
Found 7 papers, 4 papers with code
HYperbolic Self-Paced Learning for Self-Supervised Skeleton-based Action Representations
We propose to use hyperbolic uncertainty to determine the algorithmic learning pace, under the assumption that less uncertain samples should be more strongly driving the training, with a larger weight and pace.
Ranked #61 on
Skeleton Based Action Recognition
on NTU RGB+D 120
Are we certain it's anomalous?
By using uncertainty, HypAD may assess whether it is certain about the input signal but it fails to reconstruct it because this is anomalous; or whether the reconstruction error does not necessarily imply anomaly, as the model is uncertain, e. g. a complex but regular input signal.
Query-Guided Networks for Few-shot Fine-grained Classification and Person Search
Few-shot fine-grained classification and person search appear as distinct tasks and literature has treated them separately.
Class Interference Regularization
In experimental evaluation, the combination of CIR and a plain Siamese-net with triplet loss yields best few-shot learning performance on the challenging tieredImageNet.
Joint Detection and Tracking in Videos with Identification Features
Notably, our joint optimization maintains the detector performance, a typical multi-task challenge.
Knowledge Distillation for End-to-End Person Search
We employ this to supervise the detector of our person search model at various levels using a specialized detector.
Query-guided End-to-End Person Search
We extend this with i. a query-guided Siamese squeeze-and-excitation network (QSSE-Net) that uses global context from both the query and gallery images, ii.
