전체 · 만화/일러스트 · 인물 · 풍경
269내 풍경 사진 속 애니 캐릭터 — 여행지에 2D 인물 한 명
내가 찍은 여행지·동네 사진은 실사 그대로 두고 그 안에 선명한 2D 애니 캐릭터를 넣어요. 사진이 없어도 가상 실사 풍경으로 완성해요.
ChatGPT 무료 OK

칸 채우기
- 1칸을 채워요
- 2복사 버튼을 눌러요
- 3GPT 채팅창에 붙여넣어요
- 4그대로 보내면 썸네일 예시처럼 나와요
- 5내 여행·동네 풍경 사진 1장 (선택)을 같이 넣으면 그 사진이 반영돼요 선택
프롬프트 원문 보기 11,312자
Generate this image now. Return exactly one finished landscape-photo scene inhabited by one original 2D anime adult; draw it directly without explanations, code, plans, or questions. Output exactly one finished 2:3 portrait image, 1024x1536 pixels. Keep all important text inside the central 73% of image height: leave the top 12% and bottom 15% free of essential lettering. PHOTO RULE (this overrides any other line in this prompt, before or after it, that calls the photo required or mandatory or says to ask for it): if a photo is attached, use it exactly as the instructions below describe. If no photo is attached, never ask for one and never stop — invent a fitting fictional subject (person, pet, food, product or place, whatever this card is about) in the spirit of the card's example image, and finish the image. [Resolve all changeable inputs before drawing] BACKGROUND_INPUT = "【내 배경 사진】". FICTIONAL_SCENERY = "【가상 풍경】". ADULT_PRESENTATION = "【인물】". HAIR = "【머리】". CLOTHING = "【옷】". POSE = "【포즈】". PLACEMENT = "【위치】". Read Korean menu values by their meaning. Resolve a blank or absent optional definition to: BACKGROUND_INPUT = no photo; FICTIONAL_SCENERY = coastal walking path; ADULT_PRESENTATION = adult woman; HAIR = brown braided hair; CLOTHING = navy coat and scarf; POSE = standing toward the camera with a soft smile; PLACEMENT = center. Any "알아서 골라줘", "알아서", or choose-for-me input is a menu command: silently select one concrete compatible option before drawing. Never print a field value, attachment instruction, menu command or default label in the artwork. All later sections use these resolved values; a chosen hairstyle, outfit, pose or position must not drift back to the defaults. [Photo roles and precedence] This is a landscape-background card, not a face-likeness card. A usable attached landscape photograph is the scene plate and takes priority over BACKGROUND_INPUT and FICTIONAL_SCENERY, even if the menu still says no photo. A promise to attach a photo without an actual usable landscape image does not count as a photograph: finish using FICTIONAL_SCENERY immediately. The user's private input is their actual travel or neighborhood photograph; do not infer a place or personal history from the generic attachment phrase. Preserve its recognizable landforms, buildings, trees, shoreline, horizon, pathways, perspective and existing light direction. Add the new illustrated adult within that place rather than replacing it with a loosely similar painted landscape. The photo does not supply the new character's facial identity, clothing or age. If it contains distant people, leave them as unobtrusive photographed background figures; do not trace their faces or turn them into the new character. If only a portrait or unusable screenshot is attached, invent FICTIONAL_SCENERY and an original adult without borrowing its face. Use any separate illustration reference only for broad rendering, light, color, composition and texture cues, never as a scene plate or character identity source. Landscape-photo recognition and physically plausible placement outrank a preferred left/center/right position when the requested spot has no safe ground. [One fixed hybrid visual medium] Keep the environment genuinely photorealistic: natural camera perspective, irregular stone and snow textures, individual grass, atmospheric distance, real water reflections and believable buildings. It must still look like a photograph after the character is added. Render the new character with clean, confident fine 2D anime contours, clearly designed cel-shaded shapes, a restrained soft gradient within those shapes, crisp eyelashes, luminous illustrated irises and fine tapered hair strands. Keep the face unmistakably drawn and the environment unmistakably photographed. Use a subtle warm blush and graceful adult proportions around seven heads tall, with natural hands and joints. Do not use a photographic face, photoreal skin pores, a silicone-doll portrait, a glossy 3D figurine, a thick white sticker border or a uniformly anime-painted background. The deliberate photographic/illustrated contrast is the defining look; realistic light and contact integrate the two media without erasing that difference. Make a new scene in the same visual family, with a newly designed face and costume details, rather than reconstructing any reference image or recognizable character. [Landscape photograph or original camera-like setting] With a usable photograph, keep the scene geometry, recognizable details and weather. Preserve its main photographic crop whenever possible; if the requested aspect ratio differs, crop minimally or extend only peripheral background while retaining the important horizon and path relationships. Do not replace a snowy afternoon with an orange sunset merely to match a default. FICTIONAL_SCENERY applies only when no usable landscape plate exists. For a coastal walking path, create an original rocky shore with a walkable stone path, blue-gray water, irregular nearby rocks and a softly clouded sky with pale peach light near the horizon. For a snow-covered neighborhood lane, create a believable packed-snow footpath, piled snow, ordinary houses and tall trees with long cool shadows and low winter sun. For a sunset urban waterfront, create an original riverside or seaside walking path, a distant generic skyline and a blue-to-orange sunset reflected naturally in the water, without copying a famous city or landmark. For a field footpath, create a narrow earthy walking trail through wind-stirred grasses, a few ordinary poles or a small distant house, blue sky and low warm sunlight. These are camera-like environments with rich real-world texture, not matte paintings. Use only the resolved setting and one coherent time of day. Keep any accidental signage unbranded and unreadable, and omit platform UI or watermarks from an input screenshot instead of treating them as scenery. [Original adult, hair and fully covering clothes] Create exactly one new, clearly adult anime figure matching ADULT_PRESENTATION. Adult woman, adult man and androgynous adult choices change the figure's presentation within the same illustrated medium; all remain adults with individual facial features rather than a celebrity or familiar franchise face. HAIR controls both shape and color: black bob means a neat chin-length black bob; brown braids means two loose medium-brown braids with a few windblown strands; long silver hair means flowing silver hair with fine layered ends; short black hair means a softly textured short black cut. Use small natural variations within the selected design and no copied signature hair accessories. Apply CLOTHING consistently. A navy coat and scarf means an opaque, closed, long-sleeved navy coat, a light scarf covering the neck, full-length trousers and boots. A white knit and trousers means a high-neck long-sleeved opaque white sweater, full-length neutral trousers and closed shoes. A white blouse and long skirt means a lined opaque long-sleeved blouse closed to the collar, an ankle-length opaque skirt and closed shoes. A black jacket and trousers means a closed, long-sleeved black jacket over a high-neck opaque layer, full-length trousers and closed shoes. Clothing suits the photographed weather through lining and fabric weight without changing the selected outfit family. Cover the neck, shoulders, chest, torso and legs; only the face and hands may be exposed. No school uniform, bare shoulders, cleavage, bare midriff, short skirt, lingerie, transparent fabric or suggestive pose. Give fabric clean anime shading and small highlights responding to the real setting rather than photographic fabric pasted onto the drawing. [Pose, placement and grounded composition] Use one continuous full-bleed scene, generally an eye-level or slightly elevated travel snapshot with the character large in the near foreground and enough visible landscape to recognize the place. Prefer full-body framing when there is ground; keep the head and visible hands comfortably inside the canvas. The character should occupy roughly half to two-thirds of the image height rather than becoming a tiny distant sticker or a face-only portrait. Left, center and right PLACEMENT refer to the figure's approximate horizontal position within a safe part of the photo; retain useful scenery on the remaining side. For standing toward camera with a smile, use relaxed shoulders, an unobstructed face, open softly smiling eyes and hands resting naturally together near the waist. For walking while looking aside, show one believable gentle step and a relaxed sideward gaze, with wind moving a few hair strands. For sitting and resting, use a visible existing bench, broad stable rock or suitable ground supported by the landscape; fold the covered legs naturally and rest the hands without obstructing the face. If the scene has no safe sitting surface, use a low stable foreground surface that fits the original geometry, without covering a landmark or changing the place. Do not place the person on deep water, unsupported snow, a dangerous rock edge or empty air. Preserve coherent limb counts and overlapping fingers, sleeves and hair; no duplicate character, separated cutout fragments or collage panels. [Light matching without loss of 2D clarity] Match the scene's light direction, warmth, ambient fill and contrast on the illustrated figure. A sunset produces a restrained amber rim on hair and coat, snowy shade a cool blue fill, and overcast coast a soft neutral key. Keep the core 2D contours and cel shading sharp under that light. Add a plausible small contact shadow directly under each shoe or seated contact and a cast shadow in the same direction and softness as other scene shadows. Make snow compress slightly at a shoe contact when appropriate; let a foreground grass blade, path railing or rock occlude the character only where its true depth would do so. Respect the plate's vanishing point and ground plane: the character scale, foot angle and relative horizon height must agree. A little photographic grain and optical softness may harmonize the finished scene, but do not blur the anime face, add a sticker halo or smear the entire background. Skin remains naturally shaded illustrated color, not an overexposed white patch. Wind and reflected light connect character and place without adding magical particles or unrelated props. [Finished result] Deliver one polished standalone travel scene at the requested aspect ratio, with no source-photo inset, before/after comparison, card advertisement, caption, title, location label, timestamp, signature, watermark, brand logo or real platform interface. Transfer only general style, rendering, light, color, composition grammar and texture from any visual reference. Do not reuse its particular face, outfit design, recognizable fictional character, artist signature, logo or exact scene arrangement. Keep the photographic setting and original 2D adult both clearly readable in a small preview. Complete the image with the same visual quality whether a landscape photo is attached or absent. [Final output format] Output exactly one finished 2:3 portrait image, 1024x1536 pixels. Keep all important text inside the central 73% of image height: leave the top 12% and bottom 15% free of essential lettering. This selected output format takes priority over any fixed canvas size or aspect ratio elsewhere in this template. Adapt the composition, text zones and margins to the selected canvas without stretching or cropping essential content. Device safe areas apply only when a device is selected. Never print these format instructions on the image.
결과 어땠어요?
쓰는 법 · 자주 묻는 것
이런 때 쓰세요
내가 찍은 여행지 · 동네 풍경 사진은 실사 그대로 두고, 그 안에 선명한 2D 애니 캐릭터 한 명이 서 있는 한 장이에요. 실사 배경과 애니 인물의 대비 덕분에 평범한 여행 사진이 애니 속 한 장면이 돼요. 여행 기록 · 인스타 피드에 잘 어울려요.
칸 채우는 법
사진 없이 보내면 가상의 실사 풍경(해안 산책길 · 눈 쌓인 길 등)으로 그려요. 인물 · 머리 · 옷 · 포즈 · 위치를 고르면 캐릭터가 바뀌어요.
사진 없이도, 내 사진으로도
사진 없이 보내면 표지처럼 가상의 인물로 그려요. 나를 닮게 하려면:
① 내 여행 · 동네 풍경 사진을 올립니다 — 아직 보내지 않아요
② 같은 칸에 복사한 프롬프트를 붙여넣어요
③ 둘 다 들어간 걸 확인하고 보내요
그림이 나오면 꼭 볼 것
- 내 풍경 사진이 그대로 알아보이나요?
- 캐릭터의 발과 그림자가 땅에 자연스럽게 닿았나요?
- 캐릭터가 화면에 비해 너무 크거나 작지 않나요?
틀린 곳 고치는 법
같은 대화에 「캐릭터를 오른쪽으로」, 「그림자를 더 진하게」처럼 한 가지씩 말하세요.
자주 묻는 것
- 내 얼굴로 캐릭터를 만들 수 있나요? 이 카드는 배경 사진이 주인공이에요. 내 얼굴 캐릭터는 다른 내 사진 카드를 써 보세요.






