Dataset Viewer

The dataset viewer is not available because its heuristics could not detect any supported data files. You can try uploading some data files, or configuring the data files location manually.

mask-clip-db-mae-cache

Pre-computed High-Dimensional Forensic Latents for ultra-fast Diffusion-Diff forensics training.

Dataset Details

  • Source Dataset: KhangTruong/DEAL300K-30K
  • Model Extractor: mask_clip
  • Cached Channels: 14336
  • Latent Resolution: 64x64 (input 512x512)
  • Format: Parquet with zstd compression
  • Status: in_progress

Schema / Columns

  • img_id (string): Image identifier.
  • label (int64): Class label (0: authentic, 1: fully synthetic, 2: tampered/inpainted).
  • z_high_dim (binary): Pre-computed 14336-channel fp16 latent tensor bytes.
  • mask (binary): Ground-truth binary manipulation mask encoded as PNG bytes.
  • channels (int64): Total channel count (14336).
  • latent_h (int64): Latent height (64).
  • latent_w (int64): Latent width (64).

High-Speed Training with SID-UNet

sid-train --config configs/experiments/diffusion_diff_minimized/default.yaml \
          --cached-hf-repo KhangTruong/mask-clip-db-mae-cache
Downloads last month
105