Generalizable Person Re-identification
12 papers with code • 12 benchmarks • 6 datasets
Generalizable person re-identification refers to methods trained on a source dataset but directly evaluated on a target dataset without domain adaptation or transfer learning.
Instead, we aim to explore multiple labeled datasets to learn generalized domain-invariant representations for person re-id, which is expected universally effective for each new-coming re-id scenario.
Interpretable and Generalizable Person Re-Identification with Query-Adaptive Convolution and Temporal Lifting
In this paper, beyond representation learning, we consider how to formulate person image matching directly in deep feature maps.
Surpassing Real-World Source Training Data: Random 3D Characters for Generalizable Person Re-Identification
To address this, we propose to automatically synthesize a large-scale person re-identification dataset following a set-up similar to real surveillance but with virtual environments, and then use the synthesized person images to train a generalizable person re-identification model.
Domain generalization (DG) serves as a promising solution to handle person Re-Identification (Re-ID), which trains the model using labels from the source domain alone, and then directly adopts the trained model to the target domain without model updating.
In this way, human annotations are no longer required, and it is scalable to large and diverse real-world datasets.
To this end, we combine learnable batch-instance normalization layers with meta-learning and investigate the challenging cases caused by both batch and instance normalization layers.
Though online hard example mining has improved the learning efficiency to some extent, the mining in mini batches after random sampling is still limited.
In this work, we further investigate the possibility of applying Transformers for image matching and metric learning given pairs of images.
When deploying person re-identification (ReID) model in safety-critical applications, it is pivotal to understanding the robustness of the model against a diverse array of image corruptions.
The calibrated person representation is subtly decomposed into the identity-relevant feature, domain feature, and the remaining entangled one.