The paper introduces Molmo2Fish, an interactive tool that uses a multimodal large language model to correct imperfect fish tracking predictions in sonar datasets. It demonstrates that the system can effectively improve multi‑object tracking performance through natural language guidance, though further enhancements are needed. The authors provide open‑source code and data for replication.
arXiv:2608.21281v1 Announce Type: new Abstract: Recent advances in field technology have led to a massive influx of in-the-wild video data for ecological science. The primary bottleneck in leveraging...
arXiv:2610.29047v1 Announce Type: new Abstract: Understanding the properties of fish populations is increasingly important. Analyzing fish length, width, and mass is integral to maintaining healthy s...
arXiv:2609.38347v1 Announce Type: new Abstract: Quantifying collective fish behavior requires accurate trajectories, yet multi-view 3D tracking remains challenging due to frequent occlusions, visuall...
The paper presents the Huntington and Caltech Koi RGB and Sonar dataset (HaCKRS), which pairs Garmin LiveScope imaging-sonar recordings with top‑down RGB video of koi carp. The dataset includes over 105,000 temporally matched frame pairs, spanning about 3.3 hours across four range and orientation configurations, along with detailed metadata, sonar geometry, synchronization data, and planar registration for three horizontal setups. The release focuses on paired observations and calibration to aid annotation and the development of fish detection, tracking, and counting methods, and to explore the use of consumer sonar for ecological monitoring.
arXiv:2610.33478v1 Announce Type: cross Abstract: Marine video understanding is crucial for a wide range of real-world applications. However, existing marine video datasets suffer from limitations, i...
arXiv:2607. 24064v1 Announce Type: cross Abstract: Recent Vision-Language Models (VLMs) have achieved remarkable success in visual understanding, driven by the growing availability of high-quality image-text pairs.
arXiv:2609.15484v1 Announce Type: new Abstract: We report on the continued development of CatchMonitor, resulting in a prototype computer vision system designed to automatically quantify discarded fi...
FiVOS is an interactive video object segmentation algorithm tailored for fish segmentation in aquaculture. It introduces a mask block filter and serial noise filters to detect and correct erroneous propagated masks early, improving accuracy and robustness. Experiments show FiVOS achieves state‑of‑the‑art performance on two newly constructed fish‑specific datasets.
arXiv:2609.39112v1 Announce Type: new Abstract: Camera traps accumulated vast, multidimensional data for wildlife monitoring, yet translating raw media archives into meaningful ecological insights re...
arXiv:2607. 09876v1 Announce Type: cross Abstract: Automatically retrieving videos from large camera-trap datasets remains challenging.
arXiv:2606. 29357v1 Announce Type: cross Abstract: Vision-language tracking guided by natural language specifications leverages high-level semantic cues of target objects to substantially boost tracking accuracy and robustness.
arXiv:2606. 27667v1 Announce Type: cross Abstract: Artificial intelligence is transforming biodiversity monitoring by enabling automated analysis of ecological imagery collected from camera traps, drones, satellites, underwater platforms, and other sensing systems.