← V_X champion site · methodology
Pick corpora, choose any X/Y pair of metrics, get scatter + step-5 line + per-band SROCC. Per the CID22 2023 paper Table 5 band cutoffs. Data is queried in-browser with DuckDB-WASM (HTTP-range over parquet); all CPU work happens on a Web Worker.
Try first: the default pick (AIC-3 corpus, X =
q, Y = score_ssim2) shows the canonical
sweep-time encoder-quality-vs-perceptual-metric relationship.
Switch Y to score_dssim to see how DSSIM ranks pairs
differently, or to human_jnd to see how the metric
tracks the subjective JND. AIC-4 carries the same shape on a
300-row paper-curated corpus with reconstructed JND.
loading manifest…
| Band | Range (X) | n | SROCC | KROCC | PLCC | RMSE |
|---|---|---|---|---|---|---|
| no data yet | ||||||
Each box shows the Y distribution within a 5-unit X bin. Box edges = p25 / p75 (the interquartile range); whiskers = p5 / p95. For AIC-4 the reconstructed-JND CI bounds (`human_jnd_ci_lo`/`hi`) could populate the whiskers directly per-pair — TODO if Y=human_jnd.
Enter a target Y value; the table lists rows whose Y is within the tolerance, grouped by (codec, version), with their encoded bytes when available. This is the user-facing "I want zensim 70, what should the codec do?" lookup.
| codec | version | n | Y median | X median | bytes median |
|---|---|---|---|---|---|
| no query yet | |||||