DINO-Explorer: Active Underwater Discovery via Ego-Motion Compensated Semantic Predictive Coding

TL;DR

DINO-Explorer is an active perception framework for underwater monitoring that uses semantic surprise signals and ego-motion compensation to efficiently detect and prioritize significant environmental events.

Contribution

It introduces a novel, bandwidth-efficient semantic surprise-based perception system that effectively suppresses false positives and enhances event detection in autonomous underwater vehicles.

Findings

Retains 78.8% of human-verified events at a 56.8% trigger confirmation rate.

Reduces false positives by 45.5% through ego-motion compensation.

Reduces telemetry bandwidth by 48.2% while maintaining high event detection performance.

Abstract

Marine ecosystem degradation necessitates continuous, scientifically selective underwater monitoring. However, most autonomous underwater vehicles (AUVs) operate as passive data loggers, capturing exhaustive video for offline review and frequently missing transient events of high scientific value. Transitioning to active perception requires a causal, online signal that highlights significant phenomena while suppressing maneuver-induced visual changes. We propose DINO-Explorer, a novelty-aware perception framework driven by a continuous semantic surprise signal. Operating within the latent space of a frozen DINOv3 foundation model, it leverages a lightweight, action-conditioned recurrent predictor to anticipate short-horizon semantic evolution. An efference-copy-inspired module utilizes globally pooled optical flow to discount self-induced visual changes without suppressing genuine environmental novelty. We evaluate this signal on the downstream task of asynchronous event triage under variant telemetry constraints. Results demonstrate that DINO-Explorer provides a robust, bandwidth-efficient attention mechanism. At a fixed operating point, the system retains 78.8% of post-discovery human-reviewer consensus events with a 56.8% trigger confirmation rate, effectively surfacing mission-relevant phenomena. Crucially, ego-motion conditioning suppresses 45.5% of false positives relative to an uncompensated surprise signal baseline. In a replay-side Pareto ablation study, DINO-Explorer robustly dominates the validated peak F1 versus telemetry bandwidth frontier, reducing telemetry bandwidth by 48.2% at the selected operating point while maintaining a 62.2% peak F1 score, successfully concentrating data transmission around human-verified novelty events.