Oisin Mac Aodha

Affiliations

University of Edinburgh

papers

1,124

total citations

papers (26)

CleverBirds: A Multiple-Choice Benchmark for Fine-grained Human Knowledge Tracing

NEURIPS 2025arXiv

citations

Oisin Mac Aodha

Affiliations

papers (26)

The Temporal Opportunist: Self-Supervised Multi-Frame Monocular Depth

Benchmarking Representation Learning for Natural World Image Collections

Multi-Label Learning From Single Positive Labels

When Does Contrastive Visual Representation Learning Work?

Learning Stereo from Single Images

Focus on the Positives: Self-Supervised Learning for Biodiversity Monitoring

Improving Semantic Correspondence with Viewpoint-Guided Spherical Maps

Incremental Generalized Category Discovery

On Label Granularity and Object Localization

Exploring Fine-Grained Audiovisual Categorization with the SSW60 Dataset

MVSAnywhere: Zero-Shot Multi-View Stereo

Demystifying Unsupervised Semantic Correspondence Estimation

Active Learning-Based Species Range Estimation

AirPlanes: Accurate Plane Estimation via 3D-Consistent Embeddings

Virtual Occlusions Through Implicit Depth

WildSAT: Learning Satellite Image Representations from Wildlife Observations

DepthCues: Evaluating Monocular Depth Perception in Large Vision Models

ViewNet: Unsupervised Viewpoint Estimation From Conditional Generation

Representational Similarity via Interpretable Visual Concepts

The Automated LLM Speedrunning Benchmark: Reproducing NanoGPT Improvements

Visual Knowledge Tracing

SAOR: Single-View Articulated Object Reconstruction

From Coarse to Fine-Grained Open-Set Recognition

Representational Difference Explanations

CrossSDF: 3D Reconstruction of Thin Structures From Cross-Sections

CleverBirds: A Multiple-Choice Benchmark for Fine-grained Human Knowledge Tracing

papers (26)

The Temporal Opportunist: Self-Supervised Multi-Frame Monocular Depth

Benchmarking Representation Learning for Natural World Image Collections

Multi-Label Learning From Single Positive Labels

When Does Contrastive Visual Representation Learning Work?

Learning Stereo from Single Images

Focus on the Positives: Self-Supervised Learning for Biodiversity Monitoring

Improving Semantic Correspondence with Viewpoint-Guided Spherical Maps

Incremental Generalized Category Discovery

On Label Granularity and Object Localization

Exploring Fine-Grained Audiovisual Categorization with the SSW60 Dataset

MVSAnywhere: Zero-Shot Multi-View Stereo

Demystifying Unsupervised Semantic Correspondence Estimation

Active Learning-Based Species Range Estimation

AirPlanes: Accurate Plane Estimation via 3D-Consistent Embeddings

Virtual Occlusions Through Implicit Depth

WildSAT: Learning Satellite Image Representations from Wildlife Observations

DepthCues: Evaluating Monocular Depth Perception in Large Vision Models

ViewNet: Unsupervised Viewpoint Estimation From Conditional Generation

Representational Similarity via Interpretable Visual Concepts

The Automated LLM Speedrunning Benchmark: Reproducing NanoGPT Improvements

Visual Knowledge Tracing

SAOR: Single-View Articulated Object Reconstruction

From Coarse to Fine-Grained Open-Set Recognition

Representational Difference Explanations

CrossSDF: 3D Reconstruction of Thin Structures From Cross-Sections

CleverBirds: A Multiple-Choice Benchmark for Fine-grained Human Knowledge Tracing