{"id":58032,"date":"2026-10-14T10:07:40","date_gmt":"2026-10-14T02:07:40","guid":{"rendered":"https:\/\/www.1ai.net\/?p=58032"},"modified":"2026-09-20T15:04:15","modified_gmt":"2026-09-20T07:04:15","slug":"ai%e6%8f%90%e7%a4%ba%e8%af%8d%e5%88%9b%e4%bd%9c%e7%ac%ac%e5%85%ad%e5%8d%81%e4%ba%8c%e8%8a%82%ef%bc%9a%e4%b8%8d%e4%bc%9a%e5%86%99%e6%8f%90%e7%a4%ba%e8%af%8d%ef%bc%9f%e7%94%bb%e5%87%a0","status":"publish","type":"post","link":"https:\/\/www.1ai.net\/en\/58032.html","title":{"rendered":"THE FIRST SECTION OF THE AI TIP IS DEVOTED TO SECTION 62: HOW TO GENERATE A LARGE SECTION OF AI WITH HAND-DRAWN SKETCHES? CAN'T WRITE A HINT? DRAW A FEW LINES, AND IT'LL MAKE YOU UNDERSTAND"},"content":{"rendered":"<p>THE PICTURE IS ALREADY IN THE HEAD: ONE PERSON SITS IN A CHAIR AND THE OTHER IS STANDING NEXT TO HIM; OR A MAN IN A COSTUME HOLDS A LONG GUN AND THE GUN IS POINTED RIGHT AT THE CAMERA. YOU CAN REALLY OPEN AN AI TOOL WITHOUT KNOWING HOW TO WRITE THIS IMAGE AS A HINT\u3002<\/p>\n<p>WHERE'S THE CHARACTER STANDING? WHICH WAY DOES THE BODY GO? HOW HIGH IS THE HAND? HOW CLOSE IS THE GUN TO THE CAMERA? IF THEY ALL RELY ON A DESCRIPTION OF THE TEXT, THE LONGER IT IS WRITTEN, THE EASIER IT IS FOR AI TO GET IT WRONG\u3002<\/p>\n<p>In fact, when faced with such problems, we do not have to continue to study \u201chow to write this sentence\u201d. And there's a better way for White: to make a simple picture out of your mind, so that it can be presented for you\u3002<\/p>\n<p>YOU DON'T NEED TO DRAW, YOU DON'T NEED TO KNOW SKETCH. THE PERSON MAY BE A MATCHMAKER, THE CHAIR MAY BE A BOX AND THE WEAPON MAY BE A LINE. AS LONG AS THE ACTION, LOCATION, AND IMAGE ARE WELL CONNECTED, AI CAN READ FROM THESE SIMPLE LINES A LOT OF INFORMATION THAT IS DIFFICULT TO EXPLAIN\u3002<\/p>\n<p>This section, we learn how to use an abstract<a href=\"https:\/\/www.1ai.net\/en\/tag\/%e8%8d%89%e5%9b%be\" title=\"[Sees articles with [draft] labels]\" target=\"_blank\" >Draft<\/a>, WITH A FIGURE AND SCENE, CREATE AN AI IMAGE CLOSE TO YOUR OWN IDEA\u3002<\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-58033\" title=\"468e11b9j00tli4530055d000v90hkp\" src=\"https:\/\/www.1ai.net\/wp-content\/uploads\/2026\/09\/468e11b9j00tli4530055d000v900hkp.jpg\" alt=\"468e11b9j00tli4530055d000v90hkp\" width=\"1125\" height=\"632\" \/><\/p>\n<p><strong>I. TEXTS ARE MORE THAN TEXTS<\/strong><\/p>\n<p>With reference to the hints, many of the whites begin with a large text\u3002<\/p>\n<p>for example:<\/p>\n<p>A woman in a costume sitting in a chair, the sun coming out of the window, the feeling of the movie, the light..<\/p>\n<p>SUCH A TEXT WOULD CERTAINLY BE USEFUL. IT CAN TELL AI WHO, WHERE, WHAT TIME, WHAT ATMOSPHERE IT TAKES\u3002<\/p>\n<p>However, it is not very good at expressing particularly specific space relations\u3002<\/p>\n<p>FOR EXAMPLE, \u201cA WOMAN SITS ON THE RIGHT SIDE OF THE PICTURE, HAS A SLIGHT LEFT-HAND TURN, WITH A CHAIR BEHIND HER BACK, AND ANOTHER PERSON IS STANDING IN FRONT OF HER LEFT\u201d, AND EVEN IF WRITTEN IN DETAIL, AI MAY REVERSE THE RIGHT AND RIGHT POSITION OR INTERPRET THE PERSON'S MOVEMENT AS A DIFFERENT POSITION\u3002<\/p>\n<p>Then a simple sketch is more direct\u3002<\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-58035\" title=\"6502faf1j00tli45c0076d000v90jp\" src=\"https:\/\/www.1ai.net\/wp-content\/uploads\/2026\/09\/6502faf1j00tli45c0076d000v900jjp.jpg\" alt=\"6502faf1j00tli45c0076d000v90jp\" width=\"1125\" height=\"703\" \/><\/p>\n<p>Four simple sketches of action and structure<\/p>\n<p>Seeing these sketches, you might think, \"It's too simple, I can do it.\"<\/p>\n<p>That's right. That's the point of the approach\u3002<\/p>\n<p>THE SKETCHES ARE NOT FINAL, NOR ARE THEY TO PROVE YOUR PAINTING. IT'S ONLY ABOUT TELLING AI WHERE PEOPLE ARE, HOW THEY MOVE, WHERE THE PROPS POINT TO, AND HOW THE WHOLE PICTURE IS\u3002<\/p>\n<p>SO WE CAN JUST SPLIT THE MESSAGE FROM THE HINT AI INTO TWO CATEGORIES:<\/p>\n<p>TEXT TELLS AI \"WHAT IS PAINTING?\"<br \/>\nTHE PICTURE SAYS TO AI, \"WHAT DOES THIS LOOK LIKE\u3002<\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-58034\" title=\"592ba946\/j00tli45n004xd000v9000hkp\" src=\"https:\/\/www.1ai.net\/wp-content\/uploads\/2026\/09\/592ba946j00tli45n004xd000v900hkp.jpg\" alt=\"592ba946\/j00tli45n004xd000v9000hkp\" width=\"1125\" height=\"632\" \/><\/p>\n<p>The text and picture tips are each very good at expressing the message<\/p>\n<p>WHEN A CONTENT IS DIFFICULT TO ARTICULATE IN LANGUAGE, DO NOT FORCE YOURSELF TO CONTINUE WRITING. IT'S EASIER TO DRAW A RELATIONSHIP, STRAIGHT TO AI\u3002<\/p>\n<p><strong>ii. What is the sketch<\/strong><\/p>\n<p>ALTHOUGH THE SKETCHES CAN BE SIMPLE, THEY ARE NOT JUST A FEW LINES. FOR AI, IT'S NOT THE LINE THAT'S REALLY USEFUL, IT'S THE LINE THAT'S NOT PRETTY, IT'S THE LINE THAT'S SENDING A CLEAR MESSAGE\u3002<\/p>\n<p>White just needs to focus on two issues when drawing sketches\u3002<\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-58036\" title=\"b509f160j00tli45z0057d000v90hkp\" src=\"https:\/\/www.1ai.net\/wp-content\/uploads\/2026\/09\/b509f160j00tli45z0057d000v900hkp.jpg\" alt=\"b509f160j00tli45z0057d000v90hkp\" width=\"1125\" height=\"632\" \/><\/p>\n<p>Two questions to answer for a good sketch<\/p>\n<p>The first question is what the person is doing\u3002<\/p>\n<p>YOU HAVE TO DRAW THE GENERAL PHYSICAL DIRECTION OF THE PERSON, THE MOVEMENT OF HIS HANDS AND FEET AND THE LOCATION OF HIS PROPS. IF YOU WANT A GUY WITH A LONG GUN ON THE CAMERA, DRAW THE ARMS AND THE GUN'S DIRECTION. YOU CAN DRAW BIGGER AT ONE END OF THE LENS, YOU CAN DRAW SMALLER BODIES, AND IT'S EASY FOR AI TO SEE THAT WHAT YOU WANT IS AN EXAGGERATION\u3002<\/p>\n<p>The second question is: how do things fit in\u3002<\/p>\n<p>Who is on the left, who is on the right; who is close to the lens, who is far from the lens; and who is standing, sitting or being blocked by other objects. These are all \"constructions\". The image, in white, is where people and objects are placed\u3002<\/p>\n<p>IF THERE ARE MULTIPLE OBJECTS IN THE PICTURE, YOU CAN ALSO USE COLOUR TO HELP YOU DISTINGUISH AI\u3002<\/p>\n<p>For example, drawing characters in black, drawing chairs in red; or one character in black and the other in red. Colors are not here for decoration, but for labelling different elements\u3002<\/p>\n<p>HOWEVER, IT IS IMPORTANT TO NOTE THAT THE MEANING OF COLOUR IN THE SAME PICTURE DOES NOT CHANGE. IF RED STANDS FOR CHARACTER, THEN ARMS, THEN AI IS MORE CONFUSED\u3002<\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-58037\" title=\"c1308310j00tli46d005od000v90hkp\" src=\"https:\/\/www.1ai.net\/wp-content\/uploads\/2026\/09\/c1308310j00tli46d005od000v900hkp.jpg\" alt=\"c1308310j00tli46d005od000v90hkp\" width=\"1125\" height=\"632\" \/><\/p>\n<p>Distinguishing persons or props with different colours<\/p>\n<p>After the painting, you can check for yourself: if you don't read anything, can you see where this person is and what he is doing? If you don't understand yourself, you need to draw a little more clearly\u3002<\/p>\n<p><strong>iii. Why only sketches are not enough<\/strong><\/p>\n<p>The sketch solved the action and the structure, but not the other two: what the character looks like and what the scene looks like\u3002<\/p>\n<p>THE SAME MATCHMAKER, AI, CAN PAINT HER AS AN ANTIQUE WOMAN OR AS A MODERN WORKING WOMAN; THE SAME BOX MAY BE UNDERSTOOD AS EITHER A WOODEN CHAIR OR A COUCH\u3002<\/p>\n<p>SO, ONCE THE SKETCH IS COMPLETED, THERE ARE TWO TYPES OF REFERENCES THAT NEED TO BE PREPARED FOR AI\u3002<\/p>\n<p>The first is a figure set\u3002<\/p>\n<p>THE CHARACTER SET UP TO TELL AI WHO THIS CHARACTER IS. IT WOULD BE USEFUL TO HAVE A CLEAR VIEW OF THE PERSON'S FACE, HAIR, CLOTHING, COLOUR AND MAJOR ACCESSORIES. IF THERE ARE DIFFERENT ANGLES OF FRONT, SIDE, BACK, ETC., CALL THE CHARACTER FOUR. YOU DON'T NEED TO REMEMBER THIS PROFESSIONAL NAME, JUST KNOW THAT IT WORKS TO KEEP THE SAME PERSON AS POSSIBLE IN DIFFERENT IMAGES\u3002<\/p>\n<p>The second category is a scenario reference\u3002<\/p>\n<p>THE SCENE IS RESPONSIBLE FOR TELLING AI WHERE THE STORY HAPPENED. IT PROVIDES ROOM STRUCTURE, BUILDING STYLES, FURNITURE MATERIAL, LIGHT DIRECTION AND OVERALL COLOR\u3002<\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-58038\" title=\"8399fc87j00tli46v00md000v90jp\" src=\"https:\/\/www.1ai.net\/wp-content\/uploads\/2026\/09\/8399fc87j00tli46v00lmd000v900jjp.jpg\" alt=\"8399fc87j00tli46v00md000v90jp\" width=\"1125\" height=\"703\" \/><\/p>\n<p>Character set and scene reference maps<\/p>\n<p>Here, there's a clear division of labour between three types of pictures:<\/p>\n<p>The sketch is responsible for the movement of the person and the location of the image, the character chart is responsible for the character and the costume, and the scene map is responsible for the environment and light\u3002<\/p>\n<p>YOU CAN IMAGINE IT AS A MOVIE. A MAP OF THE LOCATION OF THE SCENE IS DRAWN BY THE GRASS-IMAGE DIRECTOR, A PICTURE OF THE PERSON SETTING THE IMAGE ACTOR ' S MAKE-UP, AND A REFERENCE TO THE LOCATION OF THE SCENE. THREE PLACES TOGETHER, IT'S EASIER FOR AI TO UNDERSTAND WHAT YOU REALLY WANT TO FILM\u3002<\/p>\n<p><strong>FROM AN IDEA TO AN AI IMAGE<\/strong><\/p>\n<p>Now, let's go through the whole process\u3002<\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-58039\" title=\"4fc842900tli4760053d000v90hkp\" src=\"https:\/\/www.1ai.net\/wp-content\/uploads\/2026\/09\/4fc84290j00tli4760053d000v900hkp.jpg\" alt=\"4fc842900tli4760053d000v90hkp\" width=\"1125\" height=\"632\" \/><\/p>\n<p>Full process from sketching to generating candidate images<\/p>\n<p>Suppose there's a picture in your head:<\/p>\n<p>An old man standing in a room armed with a red rim gun, pointed to the camera and had a strong film sense\u3002<\/p>\n<p>IF THAT'S ALL IT SAYS, AI KNOWS WHAT'S IN THE PICTURE, DOESN'T KNOW WHICH SIDE THE PERSON SHOULD BE ON, OR HOW FAR THE GUN GOES\u3002<\/p>\n<p>At this point, take the paper and pen, or open any software that can be painted. Using a circle for the head, with several lines for the body and limbs, and drawing the direction of the arm and the rifle. The characters do not draw five officers, nor do their clothes draw a garb\u3002<\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-58040\" title=\"a54d5282j00tli47f000vd000g00m8p\" src=\"https:\/\/www.1ai.net\/wp-content\/uploads\/2026\/09\/a54d5282j00tli47f000vd000go00m8p.jpg\" alt=\"a54d5282j00tli47f000vd000g00m8p\" width=\"600\" height=\"800\" \/><\/p>\n<p>Draw sketches<\/p>\n<p>After the sketch is completed, the figure-setting and scenario reference maps are prepared\u3002<\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-58041\" title=\"c39cf0ccj00tli47r00wud000v9000hfp\" src=\"https:\/\/www.1ai.net\/wp-content\/uploads\/2026\/09\/c39cf0ccj00tli47r00wud000v900hfp.jpg\" alt=\"c39cf0ccj00tli47r00wud000v9000hfp\" width=\"1125\" height=\"627\" \/><\/p>\n<p>Personal Four View<\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-58042\" title=\"c21203f8j00tli47z00zld000v9000hfp\" src=\"https:\/\/www.1ai.net\/wp-content\/uploads\/2026\/09\/c21203f8j00tli47z00zld000v900hfp.jpg\" alt=\"c21203f8j00tli47z00zld000v9000hfp\" width=\"1125\" height=\"627\" \/><\/p>\n<p>Personal Face Three View<\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-58046\" title=\"edb33a76j00tli486018nd000v9000hnp\" src=\"https:\/\/www.1ai.net\/wp-content\/uploads\/2026\/09\/edb33a76j00tli486018nd000v900hnp.jpg\" alt=\"edb33a76j00tli486018nd000v9000hnp\" width=\"1125\" height=\"635\" \/><\/p>\n<p>Scene 9 M.G<\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-58043\" title=\"5ff9b7f6j00tli48b00d5d000gs00ihp\" src=\"https:\/\/www.1ai.net\/wp-content\/uploads\/2026\/09\/5ff9b7f6j00tli48b00d5d000gs00ihp.jpg\" alt=\"5ff9b7f6j00tli48b00d5d000gs00ihp\" width=\"604\" height=\"665\" \/><\/p>\n<p>then we upload these pictures together to the nanobanan pro\u3002<\/p>\n<p>Add this hint and replace your drafts and ideas with:<\/p>\n<p>You're a professional photographer, and now you need to design four different extreme features, overwritten, person-acting and position-reference images I, character images II and III are my characters. Camera control: medium control, criminal control, use of Panaviment Primo Artiste, energy security, 27 millimeters, 27 millimeters<\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-58045\" title=\"0cf987bdj00tli48q00xjd000v9000hfp\" src=\"https:\/\/www.1ai.net\/wp-content\/uploads\/2026\/09\/0cf987bdj00tli48q00xjd000v900hfp.jpg\" alt=\"0cf987bdj00tli48q00xjd000v9000hfp\" width=\"1125\" height=\"627\" \/><\/p>\n<p>You'll find that you just need to change the text. Because of the most troublesome actions and locations, they have been presented in the sketches; the person's appearance and the scenery style have been given to the reference diagrams\u3002<\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-58044\" title=\"c2022913j00tli49000d2d2000v90hkp\" src=\"https:\/\/www.1ai.net\/wp-content\/uploads\/2026\/09\/c2022913j00tli49000d2d000v900hkp.jpg\" alt=\"c2022913j00tli49000d2d2000v90hkp\" width=\"1125\" height=\"632\" \/><\/p>\n<p>Case comparison from simple sketch to full image generation<\/p>\n<p>This is where the approach is really less powerful: not simply without a hint, but without having words take on all the work\u3002<\/p>\n<p><strong>V. WHY DO YOU RECOMMEND ME TO BE A QUALITY<\/strong><\/p>\n<p>AI GENERATED IMAGES BASED ON SKETCHES THAT DID NOT ALWAYS PRODUCE EXACTLY THE SAME RESULTS\u3002<\/p>\n<p>Sometimes the first person moves in the right direction, but the face is not ideal; the second person looks good, the gun goes in the wrong direction; the third light is good and the chair is in the wrong place\u3002<\/p>\n<p>This continuous process of generating and selecting results, many creators call it a \u201ctick card\u201d. It means that it is not complex, that the same requirement may have different results and that the most appropriate one needs to be selected\u3002<\/p>\n<p>At first Mr. Four Palaces will be more intuitive. Four palaces are the only four candidates. You can compare their images, characters, actions and light at the same time, select the closest one to the idea, and zoom in on it\u3002<\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-58050\" title=\"f0b0617j00tli49a00hnd000v90hkp\" src=\"https:\/\/www.1ai.net\/wp-content\/uploads\/2026\/09\/fe0b0617j00tli49a00hnd000v900hkp.jpg\" alt=\"f0b0617j00tli49a00hnd000v90hkp\" width=\"1125\" height=\"632\" \/><\/p>\n<p>First, we'll use the four-gauge to compare direction<\/p>\n<p>After getting such a four-gauge map<\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-58049\" title=\"0cf987bdj00tli49i00xjd000v9000hfp\" src=\"https:\/\/www.1ai.net\/wp-content\/uploads\/2026\/09\/0cf987bdj00tli49i00xjd000v900hfp.jpg\" alt=\"0cf987bdj00tli49i00xjd000v9000hfp\" width=\"1125\" height=\"627\" \/><\/p>\n<p>enter the following hints using the nanobanan pro model. the numbers can be changed by themselves:<\/p>\n<p>Generate a single lens image of [line 1 column 1]<\/p>\n<p>When there are advantages to each of the four images, the first option is to choose the closest to the original idea, rather than the most beautiful one\u3002<\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-58051\" title=\"61cbe7faj00tli49x003od000v90hkp\" src=\"https:\/\/www.1ai.net\/wp-content\/uploads\/2026\/09\/61cbe7faj00tli49x003od000v900hkp.jpg\" alt=\"61cbe7faj00tli49x003od000v90hkp\" width=\"1125\" height=\"632\" \/><\/p>\n<p><strong>When it is produced, don't rush to see whether it's beautiful or not<\/strong><\/p>\n<p>A lot of whites get the results, and the first thing they see is whether they're pretty or not\u3002<\/p>\n<p>BUT IN THIS WAY, THE FIRST THING THAT SHOULD BE CHECKED IS IF AI CAN READ YOUR SKETCHES\u3002<\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-58048\" title=\"9505c763j00tli4a70jtd000v90jp\" src=\"https:\/\/www.1ai.net\/wp-content\/uploads\/2026\/09\/9505c763j00tli4a700jtd000v900jjp.jpg\" alt=\"9505c763j00tli4a70jtd000v90jp\" width=\"1125\" height=\"703\" \/><\/p>\n<p>Different results obtained from sketch methods<\/p>\n<p><strong>VII. THIS METHODOLOGY IS NOT COMPLETELY WRONG<\/strong><\/p>\n<p>THE SKETCH CAN HELP AI UNDERSTAND THE IMAGE, BUT IT CANNOT BE GUARANTEED THAT IT WILL BE REPRODUCED AS IT IS\u3002<\/p>\n<p>THE MORE REFERENCES, THE BETTER. IF THERE IS A CONTRADICTION BETWEEN PERSON CHARTS, SCENE CHARTS AND SKETCHES, AI MAY NOT KNOW WHICH ONE SHOULD PREVAIL. THERE IS ALSO INTERFERENCE WITH NON-IMAGE IMAGES, SO YOU HAVE TO CLEAN UP THE MATERIAL BEFORE IT IS USED. ONLY THE FAX IS RELEVANT TO THE CURRENT IMAGE\u3002<\/p>\n<p>Complex multi-person movements, hand positions, weapon insight and face-to-face remain at risk of errors. Personalally created maps can improve the consistency of roles but do not guarantee that each created face is identical\u3002<\/p>\n<p>This section deals with common thinking, and the actual buttons and parameters still follow the tools you are using\u3002<\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-58047\" title=\"79c83f4dj00tli4ap0051d000v90hkp\" src=\"https:\/\/www.1ai.net\/wp-content\/uploads\/2026\/09\/79c83f4dj00tli4ap0051d000v900hkp.jpg\" alt=\"79c83f4dj00tli4ap0051d000v90hkp\" width=\"1125\" height=\"632\" \/><\/p>\n<p>The part where sketch control can and can still be wrong<\/p>\n<p><strong>Conclusion<\/strong><\/p>\n<p>It doesn't mean you can't make pictures in your head\u3002<\/p>\n<p>A SIMPLE SKETCH MAY BE MORE EFFECTIVE THAN A LARGE TEXT FOR THOSE DIFFICULT TO DESCRIBE: ACTION, POSITION, MASK AND STRUCTURE. THE SKETCH TELLS AI HOW TO SET IT UP, THE CHARACTER SET IT UP TO TELL AI WHO TO DRAW, THE SCENE MAP TELLS AI WHERE TO END UP WITH SHORT WORDS TO SUPPLEMENT STORIES AND EMOTIONS\u3002<\/p>\n<p>Next time you have a picture in your head and you don't know how to write a hint, don't stare at the frame\u3002<\/p>\n<p>Pick up a pen, draw a circle, add a few lines, first draw the most important positions and actions\u3002<\/p>\n<p>IT DOESN'T MATTER WHAT IT LOOKS LIKE. WHAT'S IMPORTANT IS THAT YOU FINALLY FOUND A WAY FOR AI TO SEE WHAT YOU THINK\u3002<\/p>\n<p>The final division between sketches, figures, scenes and short text<\/p>","protected":false},"excerpt":{"rendered":"<p>THE PICTURE IS ALREADY IN THE HEAD: ONE PERSON SITS IN A CHAIR AND THE OTHER IS STANDING NEXT TO HIM; OR A MAN IN A COSTUME HOLDS A LONG GUN AND THE GUN IS POINTED RIGHT AT THE CAMERA. YOU CAN REALLY OPEN AN AI TOOL WITHOUT KNOWING HOW TO WRITE THIS IMAGE AS A HINT. WHERE ARE THE PEOPLE STANDING? WHERE ARE YOU HEADED? HOW HIGH ARE YOUR HANDS? HOW CLOSE IS THE GUN TO THE CAMERA? IF THEY ALL RELY ON A DESCRIPTION OF THE TEXT, THE LONGER THEY ARE WRITTEN, THE EASIER IT IS FOR AI TO UNDERSTAND. IN FACT, WHEN FACED WITH SUCH PROBLEMS, WE DO NOT HAVE TO CONTINUE TO STUDY \u201cHOW TO WRITE THIS\u201d. AND THERE'S A BETTER WAY FOR WHITE: TO MAKE A SIMPLE PICTURE OUT OF YOUR MIND, SO THAT IT CAN BE PRESENTED FOR YOU. YOU DON'T NEED TO DRAW, YOU DON'T NEED TO KNOW SKETCH. PEOPLE CAN BE MATCHES, CHAIRS CAN BE A BOX, WEAPONS CAN BE A LINE<\/p>","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[149,144],"tags":[802,3149,956,5321,192,6481,9110],"collection":[],"class_list":{"0":"post-58032","1":"post","2":"type-post","3":"status-publish","4":"format-standard","5":"hentry","6":"category-jiaocheng","7":"category-baike","8":"tag-ai","12":"tag-prompt","13":"tag-6481","14":"tag-9110"},"acf":[],"_links":{"self":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/58032","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/comments?post=58032"}],"version-history":[{"count":0,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/58032\/revisions"}],"wp:attachment":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/media?parent=58032"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/categories?post=58032"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/tags?post=58032"},{"taxonomy":"collection","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/collection?post=58032"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}