Search Results for author:
Found 5 papers, 2 papers with code
SoccerNet 2023 Challenges Results
More information on the tasks, challenges, and leaderboards are available on https://www. soccer-net. org.
Orientation-Guided Contrastive Learning for UAV-View Geo-Localisation
We experimentally demonstrate that this prediction supports the training and outperforms previous approaches.
Ranked #1 on
Drone-view target localization
on University-1652
CLIP-ReIdent: Contrastive Training for Player Re-Identification
In this work, we investigate whether it is possible to transfer the out-standing zero-shot performance of pre-trained CLIP models to the domain of player re-identification.
Sample4Geo: Hard Negative Sampling For Cross-View Geo-Localisation
In this article, we present a simplified but effective architecture based on contrastive learning with symmetric InfoNCE loss that outperforms current state-of-the-art results.
Ranked #1 on
Image-Based Localization
on VIGOR Cross Area
Less Is More: Linear Layers on CLIP Features as Powerful VizWiz Model
Current architectures for multi-modality tasks such as visual question answering suffer from their high complexity.
