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

2026-04-14

AI Summary

This work addresses the challenge of online detection of transient oceanic events by autonomous underwater vehicles, which conventional systems struggle to identify effectively. The authors propose an active perception framework grounded in semantic surprise signals: leveraging a frozen DINOv2 model’s latent space, they employ an action-conditioned recurrent predictor to model semantic scene evolution and integrate an efference copy–inspired, optical flow–driven self-motion compensation mechanism. This novel integration enables robust disentanglement of genuine environmental novelty from visual changes induced by vehicle motion. To the best of the authors’ knowledge, this is the first approach combining self-motion compensation with foundation model–based semantic prediction for underwater novelty detection. Experimental results demonstrate that, at a fixed operating point, the method retains 78.8% of human-consensus events with a confirmation trigger rate of 56.8%, reduces false positives by 45.5% and telemetry bandwidth by 48.2% compared to a non-compensated baseline, while maintaining a peak F1 score of 62.2%.

Technology Category

• Computer Vision: Video Understanding & Activity Analysis
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Machine-in-the-loop, human agency and autonomy

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.

Problem

• active perception
• underwater monitoring
• semantic novelty
• ego-motion compensation
• autonomous underwater vehicles

Innovation

• semantic predictive coding
• ego-motion compensation
• active perception
• foundation model
• novelty detection