Search Results for author: Bharti Munjal

Found 7 papers, 4 papers with code

HYperbolic Self-Paced Learning for Self-Supervised Skeleton-based Action Representations

1 code implementation10 Mar 2023 Luca Franco, Paolo Mandica, Bharti Munjal, Fabio Galasso

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.

Action Recognition Domain Adaptation +2

Are we certain it's anomalous?

1 code implementation16 Nov 2022 Alessandro Flaborea, Bardh Prenkaj, Bharti Munjal, Marco Aurelio Sterpa, Dario Aragona, Luca Podo, Fabio Galasso

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.

Anomaly Detection Time Series +1

Class Interference Regularization

no code implementations4 Sep 2020 Bharti Munjal, Sikandar Amin, Fabio Galasso

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.

Face Verification Few-Shot Learning +2

Knowledge Distillation for End-to-End Person Search

1 code implementation3 Sep 2019 Bharti Munjal, Fabio Galasso, Sikandar Amin

We employ this to supervise the detector of our person search model at various levels using a specialized detector.

Knowledge Distillation Model Compression +3

Query-guided End-to-End Person Search

1 code implementation CVPR 2019 Bharti Munjal, Sikandar Amin, Federico Tombari, Fabio Galasso

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.

Human Detection Person Search +1

Cannot find the paper you are looking for? You can Submit a new open access paper.