Skip to content

Event

AquaProbe: Allocating LoRA Rank by Water Type for Underwater Visual Adaptation

1 reports · 1 sources

Md Shihabul Islam Shovo and co-authors propose AquaProbe, an underwater image domain adaptation method with a frozen vision model. Using paired clean images and physically rendered images, it measures the distribution of domain discrepancy across depths of the encoder, then allocates a fixed LoRA rank budget according to that distribution. The authors report that, with the same adapter parameter count of 590,000, the method achieves 70.6 mIoU on the standard SUIM split and 63.4 AUDRC under held-out degradation conditions. These values are results reported in the paper and await independent verification.

Report timeline

  1. AquaProbe: A Method for Adapting Underwater Vision Models by Allocating LoRA Rank According to Water Type

    AquaProbe proposes an underwater image adaptation method for frozen vision models: it measures how domain differences are distributed across encoder depths using paired clean images and physically rendered images, then allocates a fixed LoRA rank budget according to that distribution.

    arXiv cs.RO ·

Related topics

Perception and localizationRobot models and learningResearch papers

Interest trend

Current
3
Peak
4Nov 4 01:00
Past 24 hours
–

The trend compares participants observed throughout the same period, so its range may be smaller than the current interest count. Move or tap to inspect each hour; use the arrow keys on a keyboard.