Merging AI and underwater photography to disclose hidden ocean worlds

Within the Northeastern United States, the Gulf of Maine represents one of the crucial biologically diverse marine ecosystems on the planet — home to whales, sharks, jellyfish, herring, plankton, and tons of of other species. But whilst this ecosystem supports wealthy biodiversity, it’s undergoing rapid environmental change. The Gulf of Maine is warming faster than 99 percent of the world’s oceans, with consequences which are still unfolding.

A brand new research initiative developing at MIT Sea Grant, called LOBSTgER — short for Learning Oceanic Bioecological Systems Through Generative Representations — brings together artificial intelligence and underwater photography to document the ocean life left vulnerable to those changes and share them with the general public in latest visual ways. Co-led by underwater photographer and visiting artist at MIT Sea Grant Keith Ellenbogen and MIT mechanical engineering PhD student Andreas Mentzelopoulos, the project explores how generative AI can expand scientific storytelling by constructing on field-based photographic data.

Just because the Nineteenth-century camera transformed our ability to document and reveal the natural world — capturing life with unprecedented detail and bringing distant or hidden environments into view — generative AI marks a brand new frontier in visual storytelling. Like early photography, AI opens a creative and conceptual space, difficult how we define authenticity and the way we communicate scientific and artistic perspectives. 

Within the LOBSTgER project, generative models are trained exclusively on a curated library of Ellenbogen’s original underwater photographs — each image crafted with artistic intent, technical precision, accurate species identification, and clear geographic context. By constructing a high-quality dataset grounded in real-world observations, the project ensures that the resulting imagery maintains each visual integrity and ecological relevance. As well as, LOBSTgER’s models are built using custom code developed by Mentzelopoulos to guard the method and outputs from any potential biases from external data or models. LOBSTgER’s generative AI builds upon real photography, expanding the researchers’ visual vocabulary to deepen the general public’s connection to the natural world.

At its heart, LOBSTgER operates on the intersection of art, science, and technology. The project draws from the visual language of photography, the observational rigor of marine science, and the computational power of generative AI. By uniting these disciplines, the team just isn’t only developing latest ways to visualise ocean life — also they are reimagining how environmental stories will be told. This integrative approach makes LOBSTgER each a research tool and a creative experiment — one which reflects MIT’s long-standing tradition of interdisciplinary innovation.

Underwater photography in Latest England’s coastal waters is notoriously difficult. Limited visibility, swirling sediment, bubbles, and the unpredictable movement of marine life all pose constant challenges. For the past several years, Ellenbogen has navigated these challenges and is constructing a comprehensive record of the region’s biodiversity through the project, Space to Sea: Visualizing Latest England’s Ocean Wilderness. This huge dataset of underwater images provides the muse for training LOBSTgER’s generative AI models. The photographs span diverse angles, lighting conditions, and animal behaviors, leading to a visible archive that’s each artistically striking and biologically accurate.

LOBSTgER’s custom diffusion models are trained to duplicate not only the biodiversity Ellenbogen documents, but additionally the artistic style he uses to capture it. By learning from hundreds of real underwater images, the models internalize fine-grained details similar to natural lighting gradients, species-specific coloration, and even the atmospheric texture created by suspended particles and refracted sunlight. The result’s imagery that not only appears visually accurate, but additionally feels immersive and moving.

The models can each generate latest, synthetic, but scientifically accurate images unconditionally (i.e., requiring no user input/guidance), and enhance real photographs conditionally (i.e., image-to-image generation). By integrating AI into the photographic workflow, Ellenbogen will have the option to make use of these tools to get well detail in turbid water, adjust lighting to emphasise key subjects, and even simulate scenes that may be nearly inconceivable to capture in the sphere. The team also believes this approach may profit other underwater photographers and image editors facing similar challenges. This hybrid method is designed to speed up the curation process and enable storytellers to construct a more complete and coherent visual narrative of life beneath the surface.

In a single key series, Ellenbogen captured high-resolution images of lion’s mane jellyfish, blue sharks, American lobsters, and ocean sunfish () while free diving in coastal waters. “Getting a high-quality dataset just isn’t easy,” Ellenbogen says. “It requires multiple dives, missed opportunities, and unpredictable conditions. But these challenges are a part of what makes underwater documentation each difficult and rewarding.”

Mentzelopoulos has developed original code to coach a family of latent diffusion models for LOBSTgER grounded on Ellenbogen’s images. Developing such models requires a high level of technical expertise, and training models from scratch is a fancy process demanding tons of of hours of computation and meticulous hyperparameter tuning.

The project reflects a parallel process: field documentation through photography and model development through iterative training. Ellenbogen works in the sphere, capturing rare and fleeting encounters with marine animals; Mentzelopoulos works within the lab, translating those moments into machine-learning contexts that may extend and reinterpret the visual language of the ocean.

“The goal isn’t to exchange photography,” Mentzelopoulos says. “It’s to construct on and complement it — making the invisible visible, and helping people see environmental complexity in a way that resonates each emotionally and intellectually. Our models aim to capture not only biological realism, however the emotional charge that may drive real-world engagement and motion.”

LOBSTgER points to a hybrid future that merges direct commentary with technological interpretation. The team’s long-term goal is to develop a comprehensive model that may visualize a big selection of species present in the Gulf of Maine and, eventually, apply similar methods to marine ecosystems around the globe.

The researchers suggest that photography and generative AI form a continuum, slightly than a conflict. Photography captures what’s — the feel, light, and animal behavior during actual encounters — while AI extends that vision beyond what’s seen, toward what might be understood, inferred, or imagined based on scientific data and artistic vision. Together, they provide a strong framework for communicating science through image-making.

In a region where ecosystems are changing rapidly, the act of visualizing becomes greater than just documentation. It becomes a tool for awareness, engagement, and, ultimately, conservation. LOBSTgER continues to be in its infancy, and the team looks forward to sharing more discoveries, images, and insights because the project evolves.

For more information, contact Keith Ellenbogen and Andreas Mentzelopoulos.