readme

README.md

@@ -1,3 +1,31 @@
 ---
 license: cc
 ---
+
+# pseudo-camera-10k dataset
+
+### Work in progress (incomplete dataset)
+
+## Contents
+
+This dataset contains 10k free images from world class photographers. The images have been resized using Lanczos antialiasing, with their smaller edge shifted to 1024px.
+
+No images were upsampled in this dataset. The results of downsampling were manually validated to ensure limited artifacting.
+
+The images have been captioned with CogVLM.
+
+## Hardware
+
+These images were captioned on a single A100-80G at a rate of roughly 3.8 seconds per image.
+
+## Use cases
+
+These images may be used to train models.
+
+## Limitations
+
+Natural photos can be highly noisy. This will be learnt by the model; that may be a desired trait.
+
+The captions are not 100% accurate, but they are highly descriptive and close to 100%.
+
+The images do not reflect the multitude of diversity that much larger sets would.