CelebA

Overview edit

Variants: CelebA-wild, CelebA Unaligned, CelebA-Test, CelebA-faces, CelebA-5, CelebA Aligned, CelebA 64x64, CelebA 256x256, CelebA 128 x 128, CelebA + AFLW Unaligned, CelebA 128x128, CelebA

Associated Benchmarks

This dataset is used in 9 benchmarks:

• Image Generation - Metrics: bpd (8-bits)
• Image Inpainting - Metrics: FID, PSNR, SSIM, LPIPS
• Image Super-Resolution - Metrics: FID, PSNR, SSIM
• Interpretability Techniques for Deep Learning - Metrics: Insertion AUC score
• Multi-Task Learning - Metrics: Error
• Image Deblurring - Metrics: FID, PSNR, SSIM
• Image Colorization - Metrics: Consistency, FID
• Image Attribution - Metrics: Insertion AUC score (ArcFace ResNet-101), Deletion AUC score (ArcFace ResNet-101)
• Concept-based Classification - Metrics: Task Accuracy (%), Concept Accuracy (%)

Recent Benchmark Submissions

Research Papers

Recent papers with results on this dataset:

• EQ-CBM: A Probabilistic Concept Bottleneck with Energy-based Models and Quantized Vectors (2024) -
• Less is More: Fewer Interpretable Region via Submodular Subset Selection (2024) -
• Improving Diffusion-Based Image Synthesis with Context Prediction (2024) -
• FInC Flow: Fast and Invertible $k \times k$ Convolutions for Normalizing Flows (2023) -
• Zero-Shot Image Restoration Using Denoising Diffusion Null-Space Model (2022) -
• Making Sense of Dependence: Efficient Black-box Explanations Using Dependence Measure (2022) -
• Multi-Task Learning as Multi-Objective Optimization (2018) -
• RISE: Randomized Input Sampling for Explanation of Black-box Models (2018) -
• A Unified Approach to Interpreting Model Predictions (2017) -
• Axiomatic Attribution for Deep Networks (2017) -
• Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization (2016) -
• "Why Should I Trust You?": Explaining the Predictions of Any Classifier (2016) -
• Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps (2013) -