A poem generated from the RGB values of the image, combined with the descriptions written

...[68,60,57]　[66,58,55]　the　[67,59,56]　[70,62,59]　for　[68,60,58]　[64,57,55]　[61,54,52]　[57,52,49]　[58,53,50]　[64,59,56]　[67,62,59]　the　[63,58,55]　[64,59,56]　[62,57,54]　[57,52,49]　[51,47,44]　[49,45,43]　[48,44,43]　[46,42,41]　[47,43,42]　[50,46,46]　[51,47,48]　[50,46,47]　[48,44,45]　[46,41,43]　[41,36,39]　[36,32,36]　[35,32,35]　[37,34,36]　first　[48,44,45]　[47,43,44]　[45,41,42]　first　[44,39,41]　[45,39,41]　[47,41,43]　[52,46,48]　[52,46,48]　[49,43,45]　first　[44,37,38]　first　[52,43,44]　[53,44,45]　[59,50,51]　[59,50,51]　[56,47,48]　[53,44,45]　[54,45,46]　[56,47,48]　[57,48,49]　[60,52,52]　[60,52,51]　[62,54,52]　[61,51,50]　[62,52,51]　[64,54,53]　[57,47,46]　[53,43,42]　[59,49,48]　[63,55,52]　[66,58,55]　[67,59,56]　the　[64,56,53]　the　the　[63,55,52]　[58,50,47]　[62,54,51]　[64,56,53]　[64,56,53]　the　[63,55,52]　[62,54,51]　[64,56,52]　[69,60,55]　the　[68,59,54]　[56,47,42]　[53,44,39]　[57,48,43]　[59,50,46]　[59,49,47]　the　[62,52,50]　[58,48,46]　[53,43,41]　[48,38,36]　[47,37,35]　[53,43,41]　[60,51,48]　[61,53,50]　[60,52,49]　[58,50,47]　[56,48,45]　[56,48,45]　[57,49,46]　[60,51,48]　[60,50,48]　[59,49,47]　[53,43,41]　first　first　first　[59,49,47]　[61,51,49]　[57,47,45]　[54,44,43]　first...

"It was the first time building a snowman"

A poem generated from the RGB values of the image, combined with the descriptions written

...ribbons　[211,223,237]　on　[210,222,236]　[210,222,236]　on　[208,220,234]　on　on　on　[210,222,236]　[210,222,236]　[210,222,236]　[210,222,236]　on　[208,220,234]　on　[208,220,234]　[207,219,233]　[206,218,232]　[206,218,232]　[206,218,232]　[205,217,232]　[205,217,233]　[206,218,234]　[203,215,231]　my　my　[201,213,229]　[200,212,228]　[201,213,229]　[203,215,231]　[204,216,232]　[203,215,231]　[203,215,231]　my　my　my　my　my　[201,213,229]　my　my　[203,215,231]　[203,215,231]　[203,215,231]　my　[202,214,229]　[201,214,228]　[201,215,228]　[202,216,229]　my　[205,218,233]　[206,219,235]　[204,217,233]　[203,216,232]　[204,217,233]　[205,218,234]　[204,217,233]　my　[202,216,229]　[204,218,231]　[204,218,231]　[202,216,229]　on　[206,220,233]　[206,220,233]　[206,220,233]　[206,219,234]　[206,219,235]　[205,218,234]　[204,217,233]　[203,216,232]　[206,219,235]　on　[209,222,237]　on　[206,220,233]　[206,220,233]　[206,220,233]　[208,220,233]　[207,219,232]　[208,220,234]　[210,222,236]　[208,220,234]　[205,217,231]　[205,217,231]　[207,219,233]　[208,220,234]　[210,222,236]　on...

"Creating ribbons when playing on the snow on top of a mountain. what a fun experience. I remembered I had to hold my pee that time"

A poem generated from the RGB values of the image, combined with the descriptions written

...[57,52,39]　[58,53,40]　[63,59,46]　[71,67,54]　mom　[80,76,62]　[83,79,66]　[85,80,69]　[85,81,69]　[82,80,67]　[81,79,67]　[80,78,66]　[79,77,65]　[78,76,63]　[78,76,63]　was　[78,76,64]　[80,78,66]　[82,80,68]　[82,80,68]　[82,79,69]　[82,80,70]　[83,80,71]　[84,81,71]　[84,81,71]　[85,81,71]　[86,81,71]　[86,82,72]　mom　[87,82,74]　mom　mom　mom　mom　[88,84,74]　mom　[87,83,72]　[86,82,70]　[86,82,71]　[86,82,71]　[85,82,71]　[85,82,71]　[84,82,70]　[82,80,69]　[81,79,68]　[80,77,67]　was　[78,74,63]　mom　was　mom　[72,68,59]　mom　[74,68,60]　[76,69,60]　[77,69,60]　[77,70,60]　[78,71,60]　mom　was　[83,77,65]　[83,77,65]　[85,78,67]　[87,80,69]　[88,81,70]　[89,81,70]　[89,82,70]　[90,82,70]　[90,83,70]　[90,83,70]　[90,83,70]　[90,84,68]　[91,84,68]　[92,84,69]　[92,83,70]　[92,82,70]　[92,82,70]　[92,82,70]　[93,83,71]　[94,84,72]　[94,84,72]　[93,84,71]　[93,83,71]　[93,83,70]　[93,84,69]　[92,83,68]　[92,83,68]　[92,83,68]　[93,84,69]　[92,84,69]　[94,84,70]　[95,84,70]　[95,83,69]　[94,82,68]　[94,82,68]　[95,83,68]　[96,84,69]　[96,84,71]　[96,84,72]　mom　[97,84,72]　[96,83,71]　[95,83,71]　[95,83,71]　[96,84,72]　mom　[98,84,74]　mom　mom　mom　[98,84,72]　[99,84,72]　[99,85,72]　[99,85,72]　mom　mom　mom　mom　mom　mom　[100,86,72]　[99,85,72]　[99,85,72]　[99,85,72]　mom　mom...

"Mcdonalds breakie with mom. It was cold outside so we went in to warm up"

A poem generated from the RGB values of the image, combined with the descriptions written

...[217,226,243]　Copanhagen　[217,226,243]　[217,226,243]　Copanhagen　Copanhagen　Copanhagen　Copanhagen　Copanhagen　Copanhagen　Copanhagen　Copanhagen　Copanhagen　Copanhagen　Copanhagen　Copanhagen　Copanhagen　Copanhagen　Copanhagen　[218,226,245]　Copanhagen　Copanhagen　Copanhagen　Copanhagen　Copanhagen　Copanhagen　Copanhagen　Copanhagen　[218,225,243]　[218,225,243]　Copanhagen　Copanhagen　Copanhagen　Copanhagen　[217,226,243]　[217,226,243]　Copanhagen　[217,226,243]　[216,225,242]　[216,225,242]　[216,225,242]　[217,226,243]　[217,226,243]　[217,226,243]　Copanhagen　Copanhagen　[218,226,245]　Copanhagen　Copanhagen　Copanhagen　Copanhagen　Copanhagen　[218,226,245]　[218,227,245]　[217,226,243]　[217,226,243]　[216,225,242]　[217,226,243]　[217,226,243]　[217,226,243]　[217,226,243]　[217,226,243]　Copanhagen　Copanhagen　[216,224,243]　Copanhagen　[216,224,245]　[216,224,245]　[217,225,245]　Copanhagen　Copanhagen　Copanhagen　Copanhagen　Copanhagen　Copanhagen　Copanhagen　Copanhagen　Copanhagen　[218,226,245]　[218,226,245]　[218,226,245]　Copanhagen　[216,224,243]　[216,224,243]　[216,224,243]　[216,224,243]　[216,224,243]　Copanhagen　Copanhagen　Copanhagen　Copanhagen　Copanhagen　[216,224,243]　[216,224,243]　[216,224,243]　[216,224,243]　[216,224,243]...　

"dad and me at Copanhagen. I didn't remember that much about this picture honestly"

A poem generated from the RGB values of the image, combined with the descriptions written

...[42,99,189]　[34,91,181]　[33,88,179]　that　[36,91,182]　[40,95,186]　[36,90,181]　[33,88,180]　[30,87,181]　[23,82,176]　[20,79,173]　[13,72,166]　I　[23,79,174]　pretty　I　[27,80,173]　[35,87,179]　that　[28,79,171]　[31,82,173]　[34,85,176]　that　[38,88,178]　[35,86,175]　[32,83,172]　[32,82,171]　[32,82,171]　[35,85,174]　[39,89,178]　[37,87,176]　[30,80,169]　[33,83,172]　[32,82,171]　[31,82,171]　[32,85,173]　[38,91,179]　[43,97,185]　[35,91,178]　[19,76,163]　[31,90,176]　[32,92,178]　[33,93,180]　[34,94,182]　that　[23,82,172]　thing　[23,82,172]　[29,88,178]　[23,82,172]　[21,80,170]　[32,91,181]　[34,93,183]　[27,87,176]　[27,87,175]　[23,81,171]　thing　[27,82,175]　thing　[25,77,171]　[27,79,172]　[31,81,174]　that　[39,85,179]　[49,92,187]　[42,84,179]　[43,83,178]　[46,85,180]　that　[44,80,176]　[45,79,176]　[41,77,173]　[46,86,181]　[42,84,178]　[54,96,190]　[47,87,182]　[47,87,182]　[47,87,182]　[43,83,178]　[42,84,178]　[43,85,179]　that　[41,87,180]　[41,89,181]　[32,83,174]　remembered　[27,80,170]　[32,84,174]　[39,88,178]　[45,92,182]　[41,88,178]　[41,88,178]　[37,84,174]　[33,80,170]　[38,85,175]　[37,84,174]　[40,85,176]　[37,82,173]　[39,83,174]　[41,83,174]　how　[49,90,180]　[44,85,175]　[43,85,176]　[54,96,187]　[53,96,188]　[53,96,188]　[48,91,184]　[41,84,179]　[41,84,180]　[43,86,183]　[39,81,180]　[42,84,184]　[39,81,181]　[32,74,174]　[28,70,170]　[27,71,170]　thing　was　[41,86,185]　[36,83,180]...　

"the only thing I remembered was how pretty my mom was that day"

A poem generated from the RGB values of the image, combined with the descriptions written

...holding　holding　holding　[215,228,247]　[215,228,247]　[215,228,247]　[215,228,247]　[215,228,247]　[214,227,246]　[214,227,246]　[214,227,246]　[214,227,246]　[215,228,247]　[215,228,247]　[215,228,247]　holding　[215,228,247]　[214,227,246]　[213,226,245]　[214,227,246]　[214,227,246]　[214,227,246]　[213,226,245]　[213,226,245]　[213,227,246]　[214,228,247]　[213,228,247]　[213,228,247]　[213,228,247]　[213,228,247]　[213,228,247]　[213,228,247]　[212,227,246]　[212,227,246]　[212,227,246]　[212,227,246]　[212,227,246]　[213,228,247]　[213,228,247]　holding　holding　holding　[213,228,247]　[213,228,247]　[212,227,246]　[212,227,246]　[212,227,246]　[213,228,247]　[213,228,247]　holding　holding　holding　holding　[214,228,247]　[215,228,247]　[215,228,247]　[215,228,247]　[214,227,246]　[214,227,246]　[213,226,245]　[213,226,245]　[214,227,246]　[214,227,246]　[215,228,247]　holding　holding　holding　[215,228,247]　[215,228,247]　[215,228,247]　[215,228,247]　[215,228,247]　[215,228,247]　[215,228,247]　[215,228,247]　[215,228,247]　[215,228,247]　holding　[215,228,247]　[215,228,247]　[215,228,247]　[215,228,247]　[215,228,247]　[215,228,247]　[215,228,247]　[215,228,247]　[215,228,247]　[214,227,246]　[214,227,246]　[214,227,246]　[214,227,246]　[215,228,247]　holding　[215,228,247]　[215,228,247]　[215,228,247]　[215,228,247]　holding　[215,228,247]　[215,228,247]...　

"Me holding snow for the first time hehe"

A poem generated from the RGB values of the image, combined with the descriptions written

...[51,102,190]　[48,100,187]　[44,96,182]　[48,100,186]　big　[43,95,181]　[41,93,179]　[47,99,185]　[51,103,189]　[44,96,182]　[44,96,182]　[42,94,180]　[42,94,180]　[46,98,184]　[47,99,185]　[48,100,186]　know　[48,101,188]　[45,99,187]　[46,100,188]　[48,101,189]　[46,99,187]　[45,96,184]　[42,94,181]　[37,89,175]　even　[40,93,179]　[48,101,187]　[44,98,184]　that　a　[47,103,188]　a　[40,96,180]　a　[37,93,176]　a　[42,100,182]　[42,100,182]　[44,102,184]　[42,100,182]　[43,101,183]　[42,100,182]　[42,100,182]　[38,96,178]　that　[46,104,186]　[44,102,184]　king　[27,84,166]　[39,95,178]　[39,95,178]　[36,92,175]　[45,101,184]　know　big　[48,104,187]　[48,104,187]　[44,101,183]　[42,99,180]　[43,100,181]　[43,100,181]　[44,101,182]　[47,104,185]　big　[46,103,185]　[44,101,183]　[42,100,182]　[41,100,183]　[40,100,184]　[38,99,183]　[36,98,182]　a　even　that　[35,101,185]　a　[34,101,183]　[28,95,176]　[33,100,181]　[34,99,181]　a　[36,98,181]　[37,99,182]　[38,100,183]　[41,102,185]　[44,105,188]　[43,104,187]　[39,99,183]　[37,96,180]　[38,96,180]　[38,96,179]　[45,103,185]　[45,102,184]　[37,93,176]　[40,96,179]　[44,101,183]　[42,100,182]　a　[45,103,187]　big　[48,103,188]　[49,102,189]　[49,102,189]　[41,92,181]　king　[43,93,182]　that　[44,94,182]　[44,94,182]　[49,99,186]　big　[53,106,193]　a　[40,95,185]　[37,92,182]　a　[39,95,182]　[36,92,178]　[47,103,189]　big...　

"a very big castle that I don't even know who was the king"

A poem generated from the RGB values of the image, combined with the descriptions written

...[145,192,244]　[147,194,246]　[148,195,247]　[147,194,246]　[147,194,246]　[147,195,247]　[145,194,245]　[141,191,243]　[137,188,241]　[137,188,241]　saying　[141,192,245]　[143,194,247]　[143,195,247]　[140,193,245]　[138,191,243]　[138,191,244]　[138,191,245]　[138,191,245]　[137,190,244]　[140,193,247]　[143,196,249]　[142,195,247]　[138,191,243]　[138,191,243]　[138,191,243]　[137,190,242]　[137,190,242]　[138,191,243]　[140,193,245]　[142,195,247]　[142,196,248]　[138,194,245]　saying　[137,193,244]　[137,193,244]　saying　[138,194,245]　[136,192,243]　[137,193,244]　[137,193,244]　[137,193,244]　[140,196,246]　[143,199,248]　[143,199,248]　[140,196,245]　[138,194,243]　saying　[137,193,242]　[137,193,242]　[137,193,242]　[136,192,241]　[136,192,241]　[137,193,243]　[137,193,244]　[138,194,245]　[134,188,240]　[135,188,240]　[136,189,241]　[140,193,245]　[141,194,246]　[140,193,245]　[138,191,243]　[137,190,242]　[137,190,242]　saying　[141,194,246]　[143,196,248]　[143,196,248]　[141,194,246]　[140,193,245]　[140,193,245]　[141,194,245]　[141,194,244]　[141,194,244]　[140,193,243]　saying　saying　[141,194,244]　[142,195,245]　[141,194,244]　[141,194,244]　[142,195,245]　[142,195,245]　[145,195,246]　[146,195,247]　[145,193,247]　[146,193,247]　[140,187,241]　[142,189,243]　[143,190,245]　[140,187,243]　[138,185,241]　[140,187,243]　[141,187,245]　[142,188,246]...　

"saying hello to dolphins"

A poem generated from the RGB values of the image, combined with the descriptions written

...[175,218,253]　[175,218,253]　[176,219,253]　that　that　that　that　that　[176,219,253]　[176,219,253]　[176,219,254]　[176,219,254]　[176,219,254]　[175,217,254]　[175,217,255]　[175,217,255]　[175,217,255]　[175,217,255]　[174,216,254]　[174,216,254]　[174,216,254]　[175,217,255]　[175,217,254]　[176,217,253]　that　[178,218,254]　[179,219,255]　[180,220,255]　[181,221,255]　[180,221,255]　[180,222,255]　[178,221,255]　[178,221,255]　[181,224,255]　[182,225,255]　[179,222,254]　[178,221,253]　[178,221,252]　[178,221,251]　[178,221,252]　[176,219,253]　[172,215,250]　[173,216,251]　[176,219,254]　[175,218,253]　[175,218,253]　that　[176,219,254]　[178,221,255]　that　that　that　[178,221,255]　[178,221,255]　that　[176,219,253]　that　[178,221,252]　[178,221,254]　[178,221,255]　that　[176,219,255]　[174,218,255]　[172,216,255]　[174,218,255]　[173,217,255]　[172,216,255]　[172,216,254]　[173,217,255]　[173,217,254]　[174,218,254]　[174,218,253]　[176,221,254]　that　[178,223,255]　[179,224,255]　[179,224,255]　[180,225,255]　[179,225,255]　[178,225,255]　that　that　that　that　[176,222,255]　[176,222,255]　[175,221,255]　[174,220,254]　that　[176,220,255]　[175,219,254]　[175,219,254]...　

"that grey sweatshirt was my favorite"

A poem generated from the RGB values of the image, combined with the descriptions written

...small　small　[22,13,9]　[24,15,10]　[27,18,12]　[26,17,11]　[28,19,12]　[28,21,13]　small　[22,15,7]　[27,20,12]　[26,19,11]　[18,11,3]　small　[30,23,15]　small　stall　[42,36,28]　[40,36,27]　[41,37,28]　[44,38,30]　[46,40,31]　stall　had　[54,47,37]　[55,48,37]　[53,47,35]　stall　[52,47,35]　[57,52,40]　[55,51,39]　[52,48,36]　[55,50,38]　[57,52,40]　[58,52,40]　[58,52,40]　[57,50,39]　[55,47,36]　had　[54,46,36]　[57,48,39]　[60,51,42]　[58,49,40]　[55,46,37]　[60,51,42]　[61,52,43]　[58,49,40]　[58,49,40]　[56,47,38]　[57,48,39]　[58,49,40]　[59,50,41]　[58,49,40]　[61,52,43]　yummy　[62,53,44]　[58,49,40]　[61,52,43]　[59,50,41]　had　[62,53,44]　[56,47,38]　[62,53,44]　[66,57,48]　[59,50,41]　[59,50,41]　[62,53,44]　[59,50,41]　had　[55,46,37]　[59,50,41]　[59,50,41]　[59,51,41]　[57,50,40]　[57,50,40]　[56,49,38]　[56,50,38]　[56,50,38]　[57,51,39]　[58,52,40]　[59,52,41]　[60,53,43]　[60,53,43]　yummy　yummy　[63,56,46]　[63,56,46]　[58,51,41]　[54,47,37]　[59,52,42]　[53,46,36]...

"yummy hotdog we had from a small stall"

Image segmentation by grouping pixels based on RGB color similarity

"The image appears to consist of a combination of bright red, green, and blue patches, arranged in a pixelated fashion. Based on the colors and general structure, this could be some form of a processed image, possibly a depth map, thermal image, or an artistic effect designed to distort or highlight certain aspects of a photograph. The colors do not seem to represent a natural scene as they appear exaggerated or filtered."

ChatGPT's Image Analysis

Image segmentation by grouping pixels based on RGB color similarity

"This image appears to depict a coastal or outdoor scene segmented by RGB colors. In the center-right, there's a tall tree, possibly a palm tree, with green leaves extending outward, and its trunk is represented in red. The background is filled with a large blue area, which likely represents the sky or a large body of water, creating the impression of an open horizon. The lower part of the image is mostly red, suggesting the ground, possibly a beach or walkway. There also seems to be a small structure or object in the middle, potentially a bench, dome, or something similar, outlined in green. The overall scene evokes the feeling of a tranquil outdoor setting, perhaps near the coast or an open park.""

ChatGPT's Image Analysis

Image segmentation by grouping pixels based on RGB color similarity

"This image appears to depict a wide, open landscape, possibly a beach or desert under a clear blue sky. The upper part of the image is dominated by blue, suggesting a vast, open sky. The lower portion is mostly red, representing the ground, which could be sandy terrain or a beach. Along the horizon, green patches are visible, potentially indicating vegetation or water reflecting the sky. On the left side, there is a large structure or mound, outlined in red and green, which could be a hill, a building, or a natural feature. On the right side, there is a more triangular, tent-like object, also outlined in red, which could represent a tent or some form of geometric structure. The scene evokes a peaceful, outdoor setting, possibly a camping area or a secluded beach, with tents or natural formations in the foreground. "

ChatGPT's Image Analysis

Pixel colors in the segmented image are represented in a three-dimensional Numpy array

Pixel colors in the segmented image are represented in a three-dimensional Numpy array

Pixel colors in the segmented image are represented in a three-dimensional Numpy array

Visualization of Hugging

Visualization of Drawing

Visualization of Rubbing Hands

Visualization of Reading

Visualization of Brushing Teeth

Poem with Numbers is a website created from an exploration of image encoding using RGB values and ASCII characters. The image uploaded is encoded by extracting its RGB values, and words are also used as input to integrate them into the encoding process. These words are converted into ASCII values and matched with any RGB pixel values from the image. When a match is found, the pixel values are replaced with the text, creating a "poem with numbers." This website allows you to create a poem from any image and text you upload—whether it’s a photo tied to a cherished memory or a forgotten screenshot buried in your piles of files.

Segmenting Image is an experiment that involves partitioning images to segments based on RGB color values. If a pixel's color in the image is close enough to one of the reference colors (red, green and blue), it is recolored with the closest reference color. The segmented images are then fed to ChatGPT for image analysis. The objective is to see what the segmentation includes and what it leaves out through ChatGPT’s analysis. By looking at this way, we could see how the objects in these images are interpreted through pixels of colors.

To interact with the sketch code, press key "1", "2" and "3" to change the image. Press "w" to change the background into white, and press "b" to change it back to black. You can also drag to rotate the image using the touchpad.

Quantifying Movement is an experiment that uses an accelerometer to convert movement data into RGB values, creating visual representations of images. Meaningful interaction like hugging someone, drawing, or reading a favorite book, are recorded to explore how machines could perceive these actions through numerical color values. The objective of this experiment is inspired by my previous experiment, which is to quantify something that cannot be quantified (memories), which in this case, are our movements and interactions.

Stitching Frames is a video that aims to visualize the latent space interpolation concept in machine learning. By using frames of images as an analogy to understand the abstract idea of latent space, I created a video using the pixel sorting technique in TouchDesigner to illustrate the concept of interpolation based on my narratives. The objective of these experiments is to visualize how a machine generates meaning between data points through interpolation, highlighting the differences between this process and how we derive meaning from experiences.

WHAT IS THE LATENT SPACE?

In artificial intelligence, the latent space is a hidden, multi-dimensional space where compressed representations of data (like words, images, or sounds) are stored. AI models use this latent space to find connections and generate new meanings by blending different data points through a process known as interpolation.

While traditional interpolation involves smoothly transitioning between data points in a continuous space, I took a more metaphorical approach. By connecting different frames of a video using pixel sorting on TouchDesigner, the video shows how it could stitch one frame to another frame. Not seamlessly, but by showing how interpolation might work when pixels are sorted and arranged without smooth transitions.

A three-dimensional Numpy array is a collection of three two-dimensional arrays, each corresponding to the red, green, and blue channels. It shows how color images are represented in RGB format.