Spaces:
Running on Zero
test: a dark black-and-white night and a soft snowfall scene get no note from the picture check
Browse filesThe labelled set grew to 42 outputs (27 good). No good file fires a
condition. Two of the new ones are hard negatives, now kept as fixtures with
tests that they yield no note:
- A low-key black-and-white night (true blacks, dim highlights, no colour).
It passes two of veiled's three tests (p99 66.7 < 135, colourfulness 0.7 <
20) and is kept out by its blacks: p1 1.0 against 40. It is also now the
worst good file for flat: luma std 14.9 against 6, 2.49x clear.
- A cabin in heavy snowfall at dusk, soft by content. It fails all three
veiled tests and sits 2.01x clear of static.
veiled needs both ends of the range pulled toward grey, and no good file
has both. The three good files with lifted blacks all keep highlights of at
least 176.8 (the overcast beach, 1.31x above 135). The one good file with dim
highlights has p1 1.0. The pinned numbers in the test now say that, and the
provenance comment in outputs.py gives the 42-file counts. No threshold
moves.
|
@@ -623,13 +623,14 @@ STRUCTURE_ANALYSIS_SIZE = 256
|
|
| 623 |
#: Frames looked at per video, evenly spaced, both ends included.
|
| 624 |
STRUCTURE_SAMPLE_FRAMES = 16
|
| 625 |
|
| 626 |
-
# Thresholds from a labelled set of
|
| 627 |
-
# 2026-11-16:
|
| 628 |
# worst good and worst bad value at the analysis size. Colourfulness is
|
| 629 |
# Hasler & Süsstrunk's (15 "slightly", 33 "moderately colourful").
|
| 630 |
-
FLAT_MAX_LUMA_STD = 6.0 # luma std, 0..255 (good >=
|
| 631 |
-
|
| 632 |
-
|
|
|
|
| 633 |
VEILED_MAX_COLOURFULNESS = 20.0 # (veiled <= 13.8)
|
| 634 |
STATIC_MAX_CORRELATION = 0.10 # luma correlation at 1/8 frame (good >= 0.188, static <= 0.033)
|
| 635 |
STATIC_MIN_COLOURFULNESS = 33.0 # (static >= 62.9)
|
|
|
|
| 623 |
#: Frames looked at per video, evenly spaced, both ends included.
|
| 624 |
STRUCTURE_SAMPLE_FRAMES = 16
|
| 625 |
|
| 626 |
+
# Thresholds from a labelled set of 42 outputs, every one looked at, measured
|
| 627 |
+
# 2026-11-16: 27 good, 3 flat grey, 9 veiled, 3 colour static. Brackets: the
|
| 628 |
# worst good and worst bad value at the analysis size. Colourfulness is
|
| 629 |
# Hasler & Süsstrunk's (15 "slightly", 33 "moderately colourful").
|
| 630 |
+
FLAT_MAX_LUMA_STD = 6.0 # luma std, 0..255 (good >= 14.9, grey <= 4.3)
|
| 631 |
+
# veiled needs both: no good file has lifted blacks AND dim highlights.
|
| 632 |
+
VEILED_MIN_BLACK_LEVEL = 40.0 # luma p1 (veiled >= 57.6; dim-highlight good <= 1.0)
|
| 633 |
+
VEILED_MAX_WHITE_LEVEL = 135.0 # luma p99 (veiled <= 115.1; lifted-black good >= 176.8)
|
| 634 |
VEILED_MAX_COLOURFULNESS = 20.0 # (veiled <= 13.8)
|
| 635 |
STATIC_MAX_CORRELATION = 0.10 # luma correlation at 1/8 frame (good >= 0.188, static <= 0.033)
|
| 636 |
STATIC_MIN_COLOURFULNESS = 33.0 # (static >= 62.9)
|
|
@@ -26,6 +26,8 @@ before any statistic was measured.
|
|
| 26 |
| good_12 | local run X3a (text to audio and video), 49 frames | 24 of 49 | overcast beach with a dog: naturally low in contrast, lifted blacks, little colour |
|
| 27 |
| good_13 | local run X3a, same clip | 0 of 49 | overcast beach, first frame |
|
| 28 |
| good_14 | local run X2a (text to video), 49 frames | 24 of 49 | black-and-white noir street: no colour at all |
|
|
|
|
|
|
|
| 29 |
| bad_grey_1 | the operator's failing output, 512x960x121 | 0 of 121 | flat grey |
|
| 30 |
| bad_grey_2 | the operator's failing output | 60 of 121 | flat grey |
|
| 31 |
| bad_washed_1 | local run X4, the audio-to-video path before its fix, 704x480x113 | 0 of 113 | ship under a grey veil |
|
|
|
|
| 26 |
| good_12 | local run X3a (text to audio and video), 49 frames | 24 of 49 | overcast beach with a dog: naturally low in contrast, lifted blacks, little colour |
|
| 27 |
| good_13 | local run X3a, same clip | 0 of 49 | overcast beach, first frame |
|
| 28 |
| good_14 | local run X2a (text to video), 49 frames | 24 of 49 | black-and-white noir street: no colour at all |
|
| 29 |
+
| good_15 | local run X1 (text to video, duration head), 121 frames | 60 of 121 | low-key black-and-white night: dark, dim highlights, no colour, true blacks |
|
| 30 |
+
| good_16 | local run M04e (video to video), 121 frames | 60 of 121 | cabin in heavy snowfall at dusk: soft and low in detail by content |
|
| 31 |
| bad_grey_1 | the operator's failing output, 512x960x121 | 0 of 121 | flat grey |
|
| 32 |
| bad_grey_2 | the operator's failing output | 60 of 121 | flat grey |
|
| 33 |
| bad_washed_1 | local run X4, the audio-to-video path before its fix, 704x480x113 | 0 of 113 | ship under a grey veil |
|
|
@@ -85,6 +85,18 @@
|
|
| 85 |
"source": "evidence/local_runs/X2a_t2v_nolora/out/run-0d97868bd890/output.mp4, frame 24 of 49: black-and-white noir street: no colour at all",
|
| 86 |
"jpeg_base64": "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"
|
| 87 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 88 |
{
|
| 89 |
"name": "bad_grey_1",
|
| 90 |
"label": "bad_grey",
|
|
|
|
| 85 |
"source": "evidence/local_runs/X2a_t2v_nolora/out/run-0d97868bd890/output.mp4, frame 24 of 49: black-and-white noir street: no colour at all",
|
| 86 |
"jpeg_base64": "/9j/4AAQSkZJRgABAQAAAQABAAD/2wBDAAMCAgMCAgMDAwMEAwMEBQgFBQQEBQoHBwYIDAoMDAsKCwsNDhIQDQ4RDgsLEBYQERMUFRUVDA8XGBYUGBIUFRT/2wBDAQMEBAUEBQkFBQkUDQsNFBQUFBQUFBQUFBQUFBQUFBQUFBQUFBQUFBQUFBQUFBQUFBQUFBQUFBQUFBQUFBQUFBT/wAARCACvAQADASIAAhEBAxEB/8QAHwAAAQUBAQEBAQEAAAAAAAAAAAECAwQFBgcICQoL/8QAtRAAAgEDAwIEAwUFBAQAAAF9AQIDAAQRBRIhMUEGE1FhByJxFDKBkaEII0KxwRVS0fAkM2JyggkKFhcYGRolJicoKSo0NTY3ODk6Q0RFRkdISUpTVFVWV1hZWmNkZWZnaGlqc3R1dnd4eXqDhIWGh4iJipKTlJWWl5iZmqKjpKWmp6ipqrKztLW2t7i5usLDxMXGx8jJytLT1NXW19jZ2uHi4+Tl5ufo6erx8vP09fb3+Pn6/8QAHwEAAwEBAQEBAQEBAQAAAAAAAAECAwQFBgcICQoL/8QAtREAAgECBAQDBAcFBAQAAQJ3AAECAxEEBSExBhJBUQdhcRMiMoEIFEKRobHBCSMzUvAVYnLRChYkNOEl8RcYGRomJygpKjU2Nzg5OkNERUZHSElKU1RVVldYWVpjZGVmZ2hpanN0dXZ3eHl6goOEhYaHiImKkpOUlZaXmJmaoqOkpaanqKmqsrO0tba3uLm6wsPExcbHyMnK0tPU1dbX2Nna4uPk5ebn6Onq8vP09fb3+Pn6/9oADAMBAAIRAxEAPwD80pGLDKscj0NXdP1GaA8SFT7d6yUv1Q8puz2Boa/3cqgFAHXWt5HcZR8K56N2Ndf4PmaJLuLblAV+YDO0nI/LiuI0jwleahYfbJbkRRN91VGSRg/4V7D8LdOjU6mhTeNsaeYR7n/CgB4jV4VaFSygdR0qzpTvpTSp5ZVZozGW9QSCB+g/KtKfQGRy9sceox1NQXpgXVZgI2SHC+XubJJHrxwCaYFGe3WR2dsk7chccc0unK1qqT7AXToD04zWokIuZIm/1e1TuHUZ5wM1Ktji3eRQC27GO54NICCRReXEssMQhZiWKxk4CkdPwr234eaXrEfg+wu9Ms7m7e5k2HY+wRgM25hkbcDAwSfX1ryHRpGs5y4JjbnO3/D8K+o/hnPHeeDNIW0t3s5AJDPfQgZRfMPyBW+8cN9BkUkrAUrvwrrelWl5JgXMzh3WFpN4jXGdqjA5zznJrc+HstvrVikNwUbUo8hg/wB8pgDj2rtJNMuJxt0+5luI0X54pEy0vfrjgjngCpLDwFBqN6873ctu6gKIYE8t8Dkgkc7eeh9/WmBm6B4S/wCEfvLmM7ZvPJZCuQ2M5w3rjPWteXTyGO4hAfUdK6xNAh8wOUDSRggF85AOMipm08BgQgwPQc0AeT+LvBUOpXGm3klvLdLbh4vJiDE/OAA3TsQvGQOtJ/wr7SLTw1Pp5njvUCurpMCoUsdxHByCdw9TivWbiO0SM+c3+rccLnOeorgtQgujqlza6VpMk1jM4ke4+8GLLl3UFhkcAEnoSMAmn0A8s1a3trPxbY6Pp1vBa2QmSW6lgkBjjmKnBUnjIVQSMZ5q9420y28MWwSzv5rq4WTEhJaXyiTjeSowjAEkcevFdn4c+FdjLobxC/eazeTbcw3FsN5kDYkwwIyCMgZHQitLV7WztdCh0260+NbNGVYIWUZkwRswoBw2eP1pAeI33jm+YzK1klygARH8tkWVjzwxTDH5TzmuV1LxJrj6i2rWVrb29zZHzFjaZWllic42bVzuGB68YzXpHjW88QatFnUoLPT4rnewt0jJdXUHhWI4XAUnpkketeZS/adKiKNdXQO1VjSAO2SP4V7Dv0Jx+NMRnaz44m113ury0linBWSO48nzWIPIU9MrgYx0Oaw7ye11PSr+LWLcm9L/ACQyxtHvB+8/I7AY2571bvLBbS93/aDa6kSrx2M4Bhbcv3tx9OSfw4qjf6VPETDc+ZeQnEjSRrl0J5DZPueueaQz5d+JPgx/B3iKeAIVs5WMlu/UMmemfUVyLb9xwTxX054v8NW3jmC2s9QAt5oiyW1/0weBhhnABwP6V4J4j8K3vhi7MF5ERgZWReUkHsaAMGO8uYiSszqR6NUg1u+QMFuZV3/ewcE01otsSKQQWPNQPEPMf0FAFxdakbasyLIMfQ1dstUt1dcAIcg4kGRmsDGOSfwo3kdaAPW9B1/TriTN5DBHlWG/Yu0kjjgDis9UUsmFAAfsPevO4LyS3PyMQCOnatrSfFZtWWOdS8RI6dV/GgDqJDiWcbcgMTx9asvsSzlRyqvvUqpHJABzVOO9+2B3tduxzklTzz608wqrbpHLP1IXuaAPNQDn6VJEeeeBQy574pVXJx1oA7LQfGEenaWtrNllDHBHUf5zXsfwT8YaPe3Gp2PnlZ5gsih0wMLnv/wIV5B4I0a3uLeaW5gEo3qo3DOOa9v+D2gWEdtq+2zUMwjUOijIzuyP5flQB6RJoiTRtJCTk9CeAeK5+eymiuGlMA5XBLDIx6inXYn0q7+yWk7i1lUMVYnA6jI9KdpN9Kbaa3laSfgbFx0JPJz27UwKFvaiG3dchm7Yzlh6elCwiaN1c5hVvu5wSee/+etad/YfZyGZHiLj/VqwOenpSW+lT3LC3igVZM7wXbhfxpAVYrHExKncuNxzwRyelfTnwnsYtN8B6JczwpJakzb5rd2edZCW2gxouccHB+bn0FfOWnSY84TACRcqykcDmu+0bxhcW2k2ttFfM8dtGwEaSlPKG8nrj1PbmqsI938S6/pmm6fqcdveSHW47C4mE0c32ZZUKsAXBCgsD0I+b5a6hPiZo9jqFlbX7CzknX93K8vmhAMddq57j/GvmeTWrk6ddRRXiXbyxOJIpVbbg/eweuec59z6123hzxV4Uuo4JtaD2F5buUW1sYCIpxx95gRkZ5IP933pBc+n9G1Kz162lnsrmO5jR2jdkP8AEDg/1q01iWIAGK8E8OfEKbREt4tNsbextY5XkWBGcq5YnJO5ieh9a980HxDZeIdIGoQMvyRl5IlbLLgZIxRYZwGm+MpLKS/WVrdGednUuScdsDn2/WrOiePXudSihuRawW7khpAMbR+dcXdxpJMz7W+YlunrSmNWC4UqecccUxHrNzqmkX0iQx39s7O4GPMGSa8s8Rpb6b4su76OzkfBIjZkZw2QDuDdgCMDjj1pbJ5NNuYbuEbJVOVYjPP0qPWdSn1hibqcF24JQbAe+CR0FIDmrnXTrssjLakrHAREWIyNxBOeevHTsMV574gaGC21HzUzcgiKGBTjcSSQTjqTWtrOo3Gl+bZWEUrTbQHnhi/cpuJyQ3faO57D2rjLfRLu/v7z7TKL2SAruufvKSTjuMZCkkcdaBmVfzW2tX8k2upHLdKsUUdqibeg27ck9sVzer6O4MhsraC2kij+aNXH3dxGOpBIyMc54r1caVZ6Poy38jxSSZCW0RChiw24JwBx1ye5Ncxqnh7UrzQLrVb51v7dG+YxyjDHA+8BjPXtRck87m0eS5gW381N7nBWeIZTpk5BzkehqvLpmkQRXeg+LoPOtpIi9vNGT8r4O0+x6DPaupJmsoFjnjjRo2PlynKsMjhWIOVIB4PQ8ZqjdaLaa5a6k00oF0tuskUtw+HDgEkKf4s+n+FIo+dPiD8KL/w1ey3FnG15pYCssgO5owy7gGx1PXkDt2rzkrtLfrnrX07ci5/seURl1jjdI7q2Y8RSDgFfYgGs3xl8LdG8XXGhtaQyaVeXdpF5s6KAkjbcbtvfkdeCeaAPmxiPLXnqaayfK3rXY+MPhXrHhWaZdg1KzgOWu7QFkHA5YdV+8OTwexNceync2QaAI+uBTcdQRxTu9KyHHFAE1nfS2j7opCtd34a1y31j9y/lWtwOmT971x71wttpl1ehmgt5JFXG5lU4X0ye1dv4a0CLToGS7tllubkYSQ8+XwCAPr60AcNt7Gr1jbhXEnXHIqvHHuJJ6Ve0tsvID0UUAeieBAZdJkZ/vPMOce4/xr1v4P8Ajebwguoxiwg1KG5MZeKYY6bgMH8a8y8Mxra6BbgAh2cspH1H/wBeu38BogE/mbx8yAHaDnk0AeieJ7xL/U1nSz+wK8GfI3btpLHofStXwPp41B7+CNVeNokR93YGQfrnAHrVLxomdQt8RmNBbL5ZJyTyxz+JNHhyaTT4Na8qXD/Z87l/hIYEEe+cUAbN3oVzosepwW1t9m851H71MlCDkDB9c1iFrazljH2jzbwDMhlfBUnk7cfyFbSa7qes+HJJby+mvI4W3KJzuIGeefT5cfhU8+haVf8AiHTbmx1KHUZLpCZQ67PKPGFIIHrQBzEFgbydbiW3l8hjwoGN31Pf6Vonwxc29tvgd/Kcq3kuucc4/DtWh4hu5dD1O/V1u3k80lUEbyLEuT6fdz6+1amia9HqttYSyxYgXbjK4LYOCep/WqEylaxyNY3R+y2rEwYlIOH6df8A9Vbun2UF/E908YNs7HKI2ct03Edj9KkvrY3FrcXEbIR5T7toHTHp+v4VcjhS1mLRFVRsZjVeB7jH4UMDS0oR6daR26F2GSwaQ5bH1/Guq0bVprGMiC5ltywO8RvjI9D61x0U0inL55/LFXIZHklRd5XJOVxSA9FsZ47leBhwASCc1aXGGG3FcrpGoPbvtyHzgE11YcEY4HvQhWGthVG4FkPpWbetaPbyLcxcMpADDv7e9XpJVaPnB9BWXfhJgd77FA+/npQNGVFaWKzrbIieQI1AQcjO443ZP5+tcsrJZi9UnKsQmOgfaAP5VqXty7aioh3uFTG8jaSM5HPes9LggyvsB/eZxxjPHf8ACgB9v4bsrXy7nVYw65JMCcMVC5yMH1A5965rxZdk+DriO3QWlq0sjCFMYH3vz7Vu3uovfXrTS7XYo/A6DgcD2rkfE8ufDEox1LYNIZlR+S9hLHMm5JJijqecjaP15rn722m019Va1kLWqRKr27x/KwIPQY+U+/8AjmtR5yqvGrbik24oBz0HNZOr6/Mmo30UMeTIoR8DggDHegTOO12z+0RX2rWsjRwrIiTQSNzznGfcED86SDUvtMWhPcnEFqDEQpwdqEnn8xTPEWnX0drNOXYRvhpF/vYPB9yKy9Dt5NTuY7NQzhN8jjcAAmBnrTC53EGrWVvo63EEi/2tdyrM1qFyoVCBGDjjG319TXFnwtoMGtXmrX3hO1vmkbJhcsIt2Pm+VSAOhOK9r8C/DNPEVgl78iqYimVXJADEd88celaWlfD3T7v4W6hqM0Lz3hg8xWD/ADZAwXXk4IOc8DikB84eIvAmjXWjvqcGl6XYxjGyCAYY5PuxOetczb6fp8d1bWSaXagKpczhPmPQ4PrXpXi64n0PwrDa3dgix38iSQ3kaDaVVNp9Pm5/UV45qWtW8d95i3UwI+UBW6D8KARtm/uptBvbZ32wRSqERVxwD1JpJrfNxpgTcdu1mA9Nh5rh7zxNdSOUQkqOc9z9a09J8VXN5cRLPNsTO1N69CeMfSgZyaRD5R2qaFQsoCnAzVG1vQjYkBx610HhfSf7Y1RW5NvGwaRh6elAHoVojxaTZqPlOznFdz8Obdrm2nYhWjV4clmAPfpmuQYqlszqvAyEAPvxXXfDTEiE5/5bA5PagD1P4hg2utRBuAtrGE3Nnjk54+n61yl3430jwvb30d9dYe5hCRoqkn7w5PoDzXZ/F6NYdetVUZEtrEiBRnJ4H9TXzn4+hhe6bzpvtU0sm+OYH5DCPuhfp/8AXoA9v8P+L9M1jwndwWtyDMx3+VjDAc5OPTmlQMmg28oBUeefm28A4Tv3rxL4Yqbe+nWRmUrAxUdjkjNe0RTeZ4etQMbmdsBep5HWgDD+KGrasuuCUX8+J1Uli5DyHAHO3ggAYH410/wIvVex1gz3WJ4yDDE+STnO4D3OO9Yni3w42upPeNFJAbSDzH3NyQOwGD7frXW/DXwc/hzTLpzOJRdiKTCj7p2uQMkA9zn6UAeiXSpceHLu5FuUR1wrBTtVumM0gRSA4IjAGFFZDPJFpVxtlcBk5XJ2n8K6B9MuI4DPIFVe/Oc0wKsxaMryOenNW7SX548Ah8Ekg/SqUqNt5ydpBNXIWXMS/c+Ukt69KYjd0qAzzZ3qMYPXrXTJdsSVLHAHTPvXN2DHULgO8jSgAIA7chR0H0FM8TeNdG8C2SXet6nb6baM2xHuXxubGdo9T9KQHTSblAYep5FYniGaGUK7Axt3YNxj1NfBHiD4q6l4o8QX13Nrl3JC0jsiiVwiqSSAF7CvefgP8XNI1vQbXw7eXs1xrQndYY3ikfMZAIO/BAwd3U0WC57KFjF3HFE8kkTJ99+Swyf0rIknVftagYPmHB/Cr2l2rXWolIZFiIiLZlbCgA8k1Bc2tvax32ZvMuEn2gxjKFcckGgDHa88uTywjtkNhVHPI61heIbG7j0C5WSLaiIXy5+bn2H1710N1OlszC3jgLzKZPOmdhggfd+tcj4q1H+07fBeNWt1lTbHwzDI5b19qBEn2C3dxCweWRl3eWjBQfryB+ZrjdQKwahOUjEWUJSNSMLx7d+K0/DsZt57l8bt4VT0J5J7ms7W9QefU3lcbdsQRQB0XBAGAOT0osO5z+pQPHcX0TxlmW23ZkJJGVz/AFrnfDsEtn4qhtpCm7eYnIw69Dn2PT+VbupahNdXepSBViXyVR1ILEgLgjnoeK53SZpbXxRaNNb/AGjLAqsDkbjjjB4x1/SgSPp34Vamh8LWShzuWN1zFbvIV2sS2AAAfvf3jjNWvBunzan8LZGtbW/uLdreZGmWOOKMuqHcqszMTjI/h+lcT8HvE81vZWkAeSQyLcBYWQMjZQ/dxzuJwPwr2Lwz4qHhv4ReG5EtEFidFmbYdoWWaaYqWIBycEZ/SgbPh74o3F8+j2gkmc25BjBbcVUrgtt7clh0rxO6uH81iTu9wK+jvjBpr6d4Wlijupnt4PNBDr13CHAHHAyM/hXzlf2zQttGSWUNn6jP9aGCKXnldxDcmhJXLDk4HSmMhU9MU+BS8oUDk9BSGVSrdwT9KnstSudMmEltK0TfXg/UUIvz+tT/AGVJgMjYfWgDtNF8cJqlutpdKIrnd8jA/Ka9I8IeMLbw3o97C1kl1qUjrJE7ykJCi5JyoHJJxjntXz/JpdxAwkRS4zkMnNdZoWspfrLGWYXjIE2nqx7kUAei+MfjTq/jGEw36eXIQESWIgbV9BgA1i6LrKXWn2MVxHDOyZXZORg8kY+tYFxZSw3YjnjaKVDyjjBHSoJDLpasxKOPOSVUcZ5ByPwx/OgD1Hw14r0/w7rt5pN7aJbQOkebog7omxnDf7JyPpxXpei+KNJ1uztraxk3OkjHDptLcgfL68g1833mqHXvE8txHEqLdPt8onc3QZAruvC3g/xLbPbXtnKLaGHftXz1WUr1+7kfrQB9A6nNEljrBWNh5mnOQxzyDGDmug0i5SK3mVgVMSI5UAkABWHXtywrwTSvihIbe5hk1S5vCYmikW6jwFVhgjHOMH8K7XSvE0mtQi6jlfMyBdqHgjuKpAekGYTeG3VOZSxXb3zkcYrs10fVrmTckSJGAAA11GvOOuGbv1ryMx3c/h/5SHMkhVmLBWA4z171v6XI0Vt5R0+3nQLljI0fHPNBNz0S88G6vqDw+RDbBVGGxdwnP5NWPr13NpMrWt5by2swx+7Yhlye647VnjU1GniJtKsoIABhkeLJP1JzXMR2V3G80oWFoZJPlHmrtBPIAwevXimB3Ok6qzXiLEAFb061d8aeDdD8daMum+I9PivrVj50aSMQVbkblZSCp696r+ENDvodSSb7PIy2kys/lkK+OvBrP+Jfxs0G00qXSbGNrjVETyZryM48pxy2wnOT1GaQz5O8c/B2x8L+MNSsIL6K2tGmbyoxukKIWO0H/gOO+a92/Zx8N6d4L0PWXt9Sjv3upIyYwpV1CqeSD2y2PwryLXZILq4e6Pm7i3LPMzEkn1J960vDmux6PdF4ri4t5CpHmRPnj0w2QfypJhY+ip5HkuYwchxGQxHGax3VTM4L4wRxng1DoPiKPxDaI/2yKa6WEElRsLDpuAye/HU1V8smZvmEY6bz0pBY0SCzquc5ik4I9j/9aueuptNsfNbUEWb76eUG2sSSpB+nWtmyuXttQlaaYHbHiNlbI5zwM15p43eSXU2mNxHIroCAMg8ZHp7U0DJxr8UF2p2hIFHzbMZJBJHFZmqXqw29/qsUMt3BY2ZupRD8pAUM2MkHHK4zWNNDmLzHuIULLuwXOR+lNfXf7E8EeL4pQ7yXejXFvGoGQSwPJOewBx1qhWPMNR+KHiC3s11ebSbWGxv5ZY4XYMQ5U4YZ3ds4qlb+Prq9lthBcWcV2Nu2RVf5CqgAZJ5PH54rn/EXjU6z4I8P6EYVQ6TNcusgOS6ysrYx7EH8657QNUi0fVbG6uYjLbxTB5Ix1ZQRkflUjse4eLvjT/wjlvaweFrGW18tAxutSlWdpGYAthVVVUe3PXrXnGq/FPxF4lkSW/1eVDDapZQxxRgpHCvRFXIC+uRySTmsPXrxJ7bCZCiQlYyxYqvYZ79qxYpHbascZYjOQBmkM9o8K6ufEXgKTSbi6muy93td34bYQpA6nuv6Vz/jnTrGCRRCRGY1CBc54HT8a5vwR4qk8P38snliWIqD5Y7sPu/zxVXVry61WeSRlkmlYktsXKr7Z6UwM64VSxAkU89q2fB+npcazFJIQVj+fjnJHSsD7JMCSYW/GrFlLc6fcxzxBkkQ5BA/nSArCQhuasR3AU56mqW7DVNHtUZIzQBpwamyDl9o9BVe/uUuSJY1KSr0bpVUusuTjaB0FRvMXXHTFAHceH9T1LxeLWx8uW91CIMqOzDJjAzjnsACa0T4Q8R67C1xa6Rf3sEZ2GW3t2kQY4A3AYrza3upLaUOjMpHdTXtvwy+N154Q0s2TQwT2Eg5RraNm9+T1/HNAHn/ANlm0+/EVzDJbXUMwJVwVZT3Fdvpup+JNYnk0zSbiKNtmUeXg47nd61zfjjxkPGvjNtRSBrdZSiYfbk44zhQAPoK9I+C99DpnjjTpLiIPFI7W7AjPDKcfrimBxlt4K8Y6TK5l0S6l2Av56ZwAOrbhxivpD4X+BJfCmgR6nrAeWWd43NtbRGRI1I5Bx/Fg84r0rxXY6bc+ENauI3ijmi06bbGsgBH7tuw9a0V1yw8M29vDNezS3bxqy2tuod8Y49qGB5jdeJkuLXULS8MMaCVzDFNaAYz0BwMisa0vNGsJ4vLjjCsmHDbzn680zxjpt5qviS+vDZywrO4kCH5iDgcEipriKKeGKOPwvHEI2LF42k3t6gkk8UAW7a60KOWK7Q20boJcRkO43hVKFlIIwSf51safqDeKPEWhaetxbLBJeRblhgEYzkDJwoqjbanZJCE/wCEFsZMHq7SE/mWq94OtILfxnaak2lSaZp9qonkQysyjYMk7m6dM49qYjrf2hfiKfhroUeg6fKf7UvYfMnKOcwwZxx7sQQPZW9q+SbfxBLfSSByXLSsQW64AXj+f51rfEfxnP448SahqsxZWu5CVQtu8uMfcTPsB+ea4XQrklx82QA+CT7/AP1qTGdjc3jy6UJd3V/5YqAXewGTkDjFVySdCtlJ6q7H8xTLhsaV5g5ywX9KQGnoviW70WdVjuHSS0nMkZHZMZx7/MAefavq7wjeaZ4q8Om6Euy4mOZYkThXwMhfUd6+LfFDeXb2Vwgw8+YiPU/Kf6V9HfAbWvPZ7EOGt7iEu0bj+JSMcc84J/IUAega3pFlY5iV2kkePhwucdewrxbWbC5jSOWVSYzJ5IZhj3/xr6KMz6eHEE1yiD7scqebEeegBwR36GvLvizKLrTdOTYI33yOdi4TPygDpnuetULc8jupRNOIwBycDjj2zXWSeBdTtrOR4hFqECQnLQ4cAcnkH61T1Dw8mneItLit3FwszxkEd2yMiu/u9MZ7ScFcNsZTg47GgDwLV9KtNVsIUtNMivdR+0ozxiCJFEQwSdxIPJ4xivNtaMllfuJ7ax0+AqCkSQw+Y2W2/e64/HpmvbfCdrPb3F64Uo3lhckc9e2fpXnnjXSLa/0Ce4vZBaS2pzDcSBsKd33RgdzSGcVqHgTWrxDJCkc1uzb1KTI+MgHkqx69aSy+HWp2dw4nntoJgoOC+7aGHfb0PtXQ+FNfuvCXhi6mW53JdsXiiIOyIDIyoPIJPf0xWNqGuuDHEWPmy5kfPTnpSA0tM0HRfDUTvcW09/cg588EeWv0XGfxqlqU9uv3Gkj/ALqNk5rHu9dnKNFuO05zjvRPqXmwqzHIYDFADZhuJLMT9aiVtrcNx71XkmZSSjBh/daozdqeMEMO1AGSAe9OL8Yo3DvSBgx4xQArN0A44pq5J6UuATTlU54PNAD0iMjY6VraezQIEI43Ais9GIXmrkDGQEA5oAjmcglgcEHP0r1bwTq/katpmoLgxrNFMfThhkfzry+5tGydo3cZOK7TwIxGmCNj8yyHb7Dg00B9S+Kru71SOeLSIpI7WZSsskpVXkXGACo6Hk9+/Ndl4L8HQ+Hx9ouL2W/uGQHylVcBv+++cCs1NXtp9EhuUEcbSwowVo/l5A5wK6zQY4riGOZkheVgPmXKgAex5FAHTW1la6m6funiyvcr6+n/ANetyx8DaaV8yabYDyRIRgVhNoSXiQhxCyrx5gPfPtXWWHgHSb+ALMVOBgHzDx+tMVyI6D4Ysm2SXVsh6njn2rmPiPaeH4Ph74olt76IMumXBCpuDE+W3Tjqa72P4ceFoJdjSssgHUuSF+p6fhWpc+A/C97ZtYT/AGa4guEaKSLzNpeNhhhwfQ0mB+S1xqsZhuWm5GDhepFQaZ5lvDEgG1iOoqXx/oL+HvG+qaLtK/ZL6W22/wC45X+lTW0BjTc+N7cfhQJHS3chh0mzRsjK7Pzwf6VYtLq00rTVvNVt3nslVpBFHKImkIBwASD3Hp2qtrMuniwtYrrVoLO5Mi7YJI5TxjBJIQjjjvWJ4v8AGVgnhe90vTbm7kmkCRStKymJkVgx2DGQC2SOhx16mkUT6j8VPCmowvHd+E72WUSGWN/7U27M9RhYwK3PDf7Smk+D721uNL8ISRvbkEedqbPu4I5+Tpya8p0axsmdPtUhXcu4use7HsBkVoXdnpcAUIDK2TudlAz7Af1oA+2/hp+074f+J76LpxtJ7TX79ZWltocSQwbMkAsSDyoz07113iexttWsXg8sSZ5AYcfhXwh8J9Zj8LfEbRtTtGaJBMIZVHJ2v8pP5Gvsi98SOEQQIsjAku7cOB2AIqkKxhap4GK3lldLE58oneAxAIIxweufetOzml0yIqrtcqTh4ruNXB4/vjDfzq3aa/Cy4lu5cHpG4Yj3watTzWdwP3NwJM88/wD16AOWmm0y6mAubM2EzDqMNG3ToR3+orL8Q+DtFXSjNqDWs1tMpdIJSGDgd8Vv61DaQyGN54txXdksB+ledfFDxDZWr2mlPGGu3i8pF3hAMjHfp65oFc8Q8bPHNexwwQrbWu7EUKjAROw/rXFXzCXVpSCdoxg11HjJ0guo4BPFcyIvL27ZQfjXF3EhS43L3XA+tSUOuZQ0p+bOBzUfmZCJnoo4qrGd2STnd/k1IjFnJ6UAWN5HFRSoWbIODSB8t64qTO7NAFLBweeaYvWnsaiJ5oAfn3qSNsEZqLrTwTQBNI+enArpPh5rv9g+JrWYQwTMx8tftEYkVSehweMiuZVSykjmp7feJUbGNpB4oA+mbEwQazLFqGm6dFBdRjzlit0EU/Xn5Rww9sdRXOR/Dya31iaHSAj2jSebEHfBC85GT6YrI0jV5Z4YBFK+/b94+tdT4X1CaDVbczgyBiUO48HI4z+NAHunhyZY9HtLeQJLcQwJE74IBIwOK63SLiVAV8qMAjgB8mvPNMmkildFWMqw4AYnnjPWu70mMvb9VhBGCV+f8c4HH4UAdpp2oPG4LRiTdztZeB78D610Glf6W0htxCjk52yda47TdTlt7qKMxSOoAy4KqCc4+72Heu8sRe3MJEcMpib52w4wCP8A9XSqFY1/7Fu7i3cS24m4yFR8nGK0I/D0MCpJLbStIR91MsT7kr0rLt7DVWUO9vIyAghvOTOD29q2xcvCkX/EwltNhy3lMjhwOzErx07YoA+MP2wvgppug66PHVpdvbG/nWKTTXhOGl2ktIHJGMgZIweSTmvmeVizp8wzuHA+tfZv7ePi8XXhvwzYBFdjPNL5nGOFVeAP9418TwOzOoB78UgKfiyYarf6fH80QeZ0ODyMECsnxP4bk0G1jldmeKcHYXXk4xyD9a0vE91bJr2ntEWMUcrM524OcjPBrM8Sa/8A2vaCKRdjJjZggjbjj9MUhmdYXhMS/MAVGMGrEjiZuXwVXhW6H/PNYKlUzgsCfQ9amF/OCoL5A6A80AehfDDTP7S8ZaFYs48ye6XIB/hB3f0r7CuLae1kklhjifKshVJASAQRn8jXxn8FLqOP4raBLdxzyATAKkYJO4jC9OcZPOO1fZVwXjDvLJCw4KqiEkD0JJ/pVIlmSojuYjGJCTGSApY8NnkY/rTbm++wkRFZF4HDcY9qlmi8xN1ukbgnLKeoqhcamPK8meGOREycgAN+eOaBFS/8QJAkzSEmOMM5wcnAHb1rzDWrWG2tru/ukL392hld5PmYZBO3PoowMCvQ/EUsWm6NLeIXmggQvlGDqOPukdj0H4185+LdamuLyRRcs8RA3Dd9445NBRhXt1uy3HTA+lYtzIWYHPTmrM8u4Y7VRnYZ9xUjHW8R2Zxyc/lSE7cjnNCthWbt0Bpm7NADlbaPU1KjHv6VEB2qXPIHagCp1z7VGxwalYYqIigBy808dKYgp360ASJKVPFWor0qcFQapKakj/OgDufDviW301Ve5EjIOiRjk/nXdWHjbS9QgPk+ba3CfMqv3I5FeM2cyFXDqWI5AzVmPXJLUhhb7Y+xBz+tAH17ot5BqcdpdIPPSdVkQlzggjvXoNnqEdpZxn7I0MaEZZUZgT6DAzXg3wh8Tx6r4Vt4oz+8hcx4J5AJyP517D4e1MxeUGJO1hgbj+g6UAeg6bKLjy5l1B7c9ApUgZ+hGa7nQNTjRtv9oSOyjPyjIJ9CMVwNlvubcXHlMWChtjgAge+TgGuh0S9gt5P9Jjiht+rSY3np/skj86aA9FstY+SVWwI2GA5cAfl2qve4iswGkMiEfOilWyD0Hb9KNOt/DFjaJ5sMctu5H7tFAHPsT0qzfr4TvLZfsX9n7nAaOIgAMO4zj+RqhM+Nv20dShk1/QbC2RYxFZtMyKu3aXkI/kgr5qhUiWLnGXA/Wvdf2u5bX/hblzbW0P2aG2tYIgmc4+Xee5/v149cwabDZ2dzHcTo8fzXZkQeXGSx2bccnIHp1pWBHPeNJY5/EUCKu8RDEmB3P/1qyvE2iR6dZpJbhVhkb5s9QR6e3NbOiXGkz6utzqt1G8M9zvmRGIITIHXjt6elQ+KvEGn39lDp9qpW2i3N5kgBdmOcHIzjgDjsaQzh2jXdjcD9DVq3FvHGyOPMkZgAQcYqqygE4+72pACGBzmkB9ieCfhvH4U8R+GpItNuRI2mF31GQrlHJBCs+3qFJUc9D7V6bd6UkSvCtrkk5dS+OfU5rxzwP+09pll4F0+z1u4WS+t4PIZUjbc2DgHIGBhQv61o6z+0Z4Uh062urSO5vJ7tAzRlsmI8jaSTx0zjHpVITOvudMazTz5B9m54VpCS2OtZGoW7lzhmUEdSeo9a1nuJJrFjIqMVjVo3dAS+eSDjpWM88k9wu1Uibr8zbSP6UXEcp4xs5H0jVY42cRtC7BgeG2/Nz+VfNs05diSdxr63ubiUrJHLEl0pQoxUhuCD1r518dfDu80LUi9haTXNjMSY/LUuyf7Jx/PvSY7HEyPxnOKqE7mx1zVm8tbi0cpPBJCw/hkUqf1qqn3sngUhkrt8oUdqbjpS5B6imnvQAu7JqdD8oqAGnqeevFADWXB+tQsNv0qd+CagY9aABTThUYPpUmePegBafu96jzQCSaAL1rcRW2WClmPrT22xMZAhSJ/Q5H4iqAOKmW4x15AoA9I+C+qyab4qe2jKtBdR5CvkruH09s19N+HjOisUmRvNG0orkK2exJFfL3wJinm8e2kVvD9obZKQu0HA2HJ59q+mdMaSWQESidgRuSI7c9zjAINMD1HTr0yH50eYIuw49MdOwro9Fg8sRmMErkhTI6lVB55Axk1w+k6tqE88Bt4PMgCHfAt0mWPdSuMj866bTLud7/ZHocUQVSZmt50llLEfKACRtGCTyKAO7tNBc24H9oWsKAZRIwAQBnjqeDxVvSbazulS3K2aLCn7uM7gVI4PIbjPFcto/iaxtkuTc6PdxrETlRGGIwTk9cdB2z1roZNUvLpYHs/DdxfWTj968MkYZevo/X8KYj4V/aWmeT4v68sjBnjdIyykkErGo7/SvEvEkkhuEW33EtEqnZnmvs/4pfswXPj7xDfa5od2bCWd91zYX4EggfAyBIjMfwK/jXx62qWi3GwuyhWwflpMEc5deHNRs40aaER7gCAWFLceHr6OyEzRDaMkkMOlfUum/sqeJvG+iaRdJfWWnSXsKTRQXqyKdjDcp3gFclSDtJzU3i79lx/CHhbUW1XVHurmHTZrhI7RFCrKgBCsdxJUjPOB0pAfHpBwOMV2Xgb4Ta/8QhKdHS3cxKHYSzBMAnHf3rmlSFHliljYyHhOcbT719mfAnwAPAnhLN/bWlxfXan/AEqCSRwY2AKq3y4BBPOKBnz9cfsweNoc7o7EEDcVF0M4/Ksm4+C3irThFE1lHK2/P7qdWx7nmvs/VPDMzmNC8Vs2PlRJDgjvjOM1yl5pU0dw4S6lsriLO1vO5Yewzjv0qrE3MXLxWUP7+NtiLujYkZ6d+hqvbwEs8ksoVX7l+TTJdFW4uHAYwFVAZInAV+OW6YGfSqF3p0MBVGvXEnOFdh83HVf0/KlYEK0M7S4gVmXHzFTnH144qpNCyjzBDJIAPmdTnH1x0qhdlLcSCZpCw6OvKkY9s1ni6SJmgeYgtt27srjr1/SgbPNviQfP1a5uJX82QDCqOoCsF+b+mPSuElXb+J6VreJrh7vVbyYM7AStuDHOMseKzpWywGcjaDwKQEGcUwvUjpnOKayEd6BiKeBTxSbcAU8KQKAGvzntmoWqV8ZI9KhfvQAzdT1Y96rsecD86kQ496AJd1KDUZozQBIzU3d1+lJ1phOKAPU/gU9pFrN5JdyNCpgKRyiFZQGyDgqxA5x6179Z68scCb4YpACFJKBA3PHCmvmvwg39mHTIssHnnWRivYE4r6C0+bU5JCNNKTXJkJQMBEVj4wN3POAcn6dKaEz0fRnv2I88FEdgojjUKCD67lI9AK6vwzoF3/pM9jI8Ylc7kV0O0nscDoOO1cPoPiBbyMx3UgnKgecuxnVvb5iOM11mk6xEl2Xt/taWiJ5clspRUkGDgY7AZzwaYrnX6L4P1eC+kvp5reRnGWJlVW/76Vc/nXSWtjcaVaBZdRna+LB4LZ3RkLehYKM4x1auFh8ceGNKQRPpoht5GzKELNlu+QR2rp0+ImlpFAiMUNwQIVeItuzjHOOByOtAzejj/wCJl5t3cxXBbaZYhJwnHORkDvX5y2/gaPxP8epfDFkoktbnXHtQbTDqITMQWXbkYCZORxgV9z/EGfUD4K8QRwXcEN89m7IYoTGBkEg55ycDHX8q+IfhjcN4b+O/hlba5ErpqEESyFSF+cgHPGejemakZ+mTjTdPsoLUW+Y4QIooUjyFUDAGccYHrXlvxg1/QvC3hjU7rVpI18O3sR064ELEurSBhuO1cjA/Wuntje3GkiWa+EiSoSFnUvg4zyBjI/GvBf2u/Gek+GPhymiNZW93qGvBH3mFgECnIkBzww6Ac9TRcR8NarcRPrbyxOZIxIMSHq2O/wClfeXgn4n6d8SfDl3NZp50dlIIBE67FY7RyrY4OCa/P6TAbjnmvp39lvx/psWjT+HnVLK+SQyo8cbN9ozjLOeR8o4xx29KaBnsmoWrXd1veGR40+eIm4I2gnGAd3X+lY0+pNcXlvG1hDBMjbEl+1qGLE42nAOfr7Vb1jUoJWuLeTC/ZWUSIgYbDjcrAg8+vSssPDrbyoV3IFG2RmY5zkDjjB496GxFLURNp5MzeZOhL7lS4JZcjjG4dufrXLX+qI8pR4DGrjdGZpRkj8q2b6bdI9rdZcquPMVuDzxkYzn6HpXOa1LHprMttM4LN/q24APckgZP0yKAK969h5Ms6QGI5AYCEpyeemeRWCz+WGd1eSJl3fvF5VR0IxV2W7e6kJWXLKSMqG59jk1zPiGaW1tbp2YZEbFUySOh7/hTA8wuI2uLSWcEnzp+fT2/nVF3HnNj7o4FbUkCWmjwF5My+X5nlgcYPTn/AD0rAqSh+QvNMLZPqaC3FMU5NAEmc8VMuCPWq4b5qlQnPpQB/9k="
|
| 87 |
},
|
| 88 |
+
{
|
| 89 |
+
"name": "good_15",
|
| 90 |
+
"label": "good",
|
| 91 |
+
"source": "evidence/local_runs/X1_duration_head/out/run-e1b342c9e31d/output.mp4, frame 60 of 121: low-key black-and-white night: dark, dim highlights, no colour, true blacks",
|
| 92 |
+
"jpeg_base64": "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"
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"name": "good_16",
|
| 96 |
+
"label": "good",
|
| 97 |
+
"source": "evidence/local_runs/M04e_v2v_s00/out/run-0df36a197101/output.mp4, frame 60 of 121: cabin in heavy snowfall at dusk: soft and low in detail by content",
|
| 98 |
+
"jpeg_base64": "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"
|
| 99 |
+
},
|
| 100 |
{
|
| 101 |
"name": "bad_grey_1",
|
| 102 |
"label": "bad_grey",
|
|
@@ -236,6 +236,28 @@ def test_a_bright_overcast_picture_is_not_called_washed_out(name):
|
|
| 236 |
assert check.conditions == () and check.notes == ()
|
| 237 |
|
| 238 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 239 |
def test_a_black_and_white_picture_yields_no_note():
|
| 240 |
real = load("good_14")
|
| 241 |
frame = load("good_1")
|
|
@@ -391,12 +413,16 @@ def test_the_check_needs_neither_torch_nor_a_media_stack():
|
|
| 391 |
# thresholds, and the numbers that put them there
|
| 392 |
# ---------------------------------------------------------------------------
|
| 393 |
|
| 394 |
-
#
|
| 395 |
-
WORST_GOOD_LUMA_STD =
|
| 396 |
WORST_GREY_LUMA_STD = 4.30
|
| 397 |
WORST_VEILED_BLACK_LEVEL = 57.6
|
| 398 |
-
WORST_GOOD_WHITE_LEVEL = 155.0
|
| 399 |
WORST_VEILED_WHITE_LEVEL = 115.1
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 400 |
WORST_VEILED_COLOURFULNESS = 13.8
|
| 401 |
WORST_GOOD_CORRELATION = 0.188
|
| 402 |
WORST_STATIC_CORRELATION = 0.033
|
|
@@ -404,10 +430,11 @@ WORST_STATIC_COLOURFULNESS = 62.9
|
|
| 404 |
|
| 405 |
|
| 406 |
def test_the_thresholds_sit_between_the_measured_classes_with_room_either_side():
|
| 407 |
-
assert WORST_GREY_LUMA_STD * 1.3 < outputs.FLAT_MAX_LUMA_STD < WORST_GOOD_LUMA_STD /
|
| 408 |
assert outputs.VEILED_MIN_BLACK_LEVEL * 1.4 < WORST_VEILED_BLACK_LEVEL
|
|
|
|
| 409 |
assert WORST_VEILED_WHITE_LEVEL * 1.15 < outputs.VEILED_MAX_WHITE_LEVEL
|
| 410 |
-
assert outputs.VEILED_MAX_WHITE_LEVEL * 1.
|
| 411 |
assert WORST_VEILED_COLOURFULNESS * 1.4 < outputs.VEILED_MAX_COLOURFULNESS
|
| 412 |
assert WORST_STATIC_CORRELATION + 0.05 < outputs.STATIC_MAX_CORRELATION
|
| 413 |
assert outputs.STATIC_MAX_CORRELATION + 0.05 < WORST_GOOD_CORRELATION
|
|
|
|
| 236 |
assert check.conditions == () and check.notes == ()
|
| 237 |
|
| 238 |
|
| 239 |
+
def test_a_dark_black_and_white_night_with_true_blacks_yields_no_note():
|
| 240 |
+
"""Dim highlights and no colour, like a veiled frame, and a narrow brightness spread; what
|
| 241 |
+
it has that a veil takes away is true blacks."""
|
| 242 |
+
check = outputs.assess_picture(stack(load("good_15")))
|
| 243 |
+
measured = check.measurements
|
| 244 |
+
assert measured.white_level < outputs.VEILED_MAX_WHITE_LEVEL
|
| 245 |
+
assert measured.colourfulness < outputs.VEILED_MAX_COLOURFULNESS
|
| 246 |
+
assert measured.black_level < outputs.VEILED_MIN_BLACK_LEVEL / 10
|
| 247 |
+
assert measured.luma_std > outputs.FLAT_MAX_LUMA_STD * 2
|
| 248 |
+
assert check.conditions == () and check.notes == ()
|
| 249 |
+
|
| 250 |
+
|
| 251 |
+
def test_a_soft_snowfall_scene_yields_no_note():
|
| 252 |
+
"""Soft and low in detail by content: kept out of every condition, not just one."""
|
| 253 |
+
check = outputs.assess_picture(stack(load("good_16")))
|
| 254 |
+
measured = check.measurements
|
| 255 |
+
assert measured.black_level < outputs.VEILED_MIN_BLACK_LEVEL
|
| 256 |
+
assert measured.white_level > outputs.VEILED_MAX_WHITE_LEVEL
|
| 257 |
+
assert measured.long_range_correlation > outputs.STATIC_MAX_CORRELATION
|
| 258 |
+
assert check.conditions == () and check.notes == ()
|
| 259 |
+
|
| 260 |
+
|
| 261 |
def test_a_black_and_white_picture_yields_no_note():
|
| 262 |
real = load("good_14")
|
| 263 |
frame = load("good_1")
|
|
|
|
| 413 |
# thresholds, and the numbers that put them there
|
| 414 |
# ---------------------------------------------------------------------------
|
| 415 |
|
| 416 |
+
# Measured by this check on the labelled dataset (42 files: 27 good, 3 grey, 9 veiled, 3 static).
|
| 417 |
+
WORST_GOOD_LUMA_STD = 14.93
|
| 418 |
WORST_GREY_LUMA_STD = 4.30
|
| 419 |
WORST_VEILED_BLACK_LEVEL = 57.6
|
|
|
|
| 420 |
WORST_VEILED_WHITE_LEVEL = 115.1
|
| 421 |
+
# veiled needs both ends of the range pulled in, so a good file is kept out by either end:
|
| 422 |
+
# good files with lifted blacks (p1 >= 40) all keep highlights of at least this (the beach),
|
| 423 |
+
GOOD_LIFTED_BLACKS_MIN_WHITE_LEVEL = 176.8
|
| 424 |
+
# and good files with dim highlights (p99 < 135) all keep blacks of at most this (the night).
|
| 425 |
+
GOOD_DIM_HIGHLIGHTS_MAX_BLACK_LEVEL = 1.0
|
| 426 |
WORST_VEILED_COLOURFULNESS = 13.8
|
| 427 |
WORST_GOOD_CORRELATION = 0.188
|
| 428 |
WORST_STATIC_CORRELATION = 0.033
|
|
|
|
| 430 |
|
| 431 |
|
| 432 |
def test_the_thresholds_sit_between_the_measured_classes_with_room_either_side():
|
| 433 |
+
assert WORST_GREY_LUMA_STD * 1.3 < outputs.FLAT_MAX_LUMA_STD < WORST_GOOD_LUMA_STD / 2.4
|
| 434 |
assert outputs.VEILED_MIN_BLACK_LEVEL * 1.4 < WORST_VEILED_BLACK_LEVEL
|
| 435 |
+
assert GOOD_DIM_HIGHLIGHTS_MAX_BLACK_LEVEL * 10 < outputs.VEILED_MIN_BLACK_LEVEL
|
| 436 |
assert WORST_VEILED_WHITE_LEVEL * 1.15 < outputs.VEILED_MAX_WHITE_LEVEL
|
| 437 |
+
assert outputs.VEILED_MAX_WHITE_LEVEL * 1.3 < GOOD_LIFTED_BLACKS_MIN_WHITE_LEVEL
|
| 438 |
assert WORST_VEILED_COLOURFULNESS * 1.4 < outputs.VEILED_MAX_COLOURFULNESS
|
| 439 |
assert WORST_STATIC_CORRELATION + 0.05 < outputs.STATIC_MAX_CORRELATION
|
| 440 |
assert outputs.STATIC_MAX_CORRELATION + 0.05 < WORST_GOOD_CORRELATION
|