Aqbeż għall-kontenut ewlieni
Malti— Agħżel lingwa
Menu

Prompt viżwali avvanzati: kompożizzjoni, dawl, perspettiva

Lezzjoni 9 / 12 Aġġornat 2 ta’ Novembru 2026 Zviluppatur
F'din il-paġna
  1. Prompt viżwali avvanzati: kompożizzjoni, dawl, perspettiva
  2. Anatomija ta’ prompt viżwali professjonali
  3. Tgħallim tal-kompożizzjoni
  4. Tgħallim tad-dawl
  5. Tgħallim tal-perspettiva
  6. Builder tal-prompt riutilizzabbli

Aġġornament 2 ta’ Novembru, 2026 — mudelli tal-immaġni. gpt-image-1, li għalih kien miktub oriġinarjament dan il-kors, ġie interrott fil-23 ta’ Ottubru, 2026, u gpt-image-1.5, gpt-image-1-mini, u chatgpt-image-latest se jiġu rtirati fl-1 ta’ Diċembru, 2026. L-eżempji issa jużaw gpt-image-2.5-sunburst, l-aktar mudell preċiż għall-editjar; gpt-image-2.5-flare huwa aktar mgħaġġel għall-ġenerazzjoni standard. It-tnejn jappoġġjaw ukoll is-settings tal-kwalità xhigh u max. Il-mudelli GPT Image dejjem jirritornaw immaġni f’base64: il-parametru response_format minn mudelli DALL·E aktar antiki mhuwiex aċċettat u tneħħa mill-kodiċi. Sors: Image generation (OpenAI).

Prompt viżwali avvanzati: kompożizzjoni, dawl, perspettiva

Prompt strutturat tajjeb huwa dak li jifred immaġni deċenti minn waħda ta’ kwalità professjonali. Il-lever mhuwiex it-tul—huwa l-vokabolarju. GPT Image jifhem it-terminoloġija tal-fotografija, taċ-ċinema u tal-arti viżwali. Li titkellem dik il-lingwa huwa bħal li tagħti direzzjonijiet lid-direttur tal-fotografija minflok ma titlob xewqa. Din il-lezzjoni tgħallmek dak il-vokabolarju u kif torganizzah.

Anatomija ta’ prompt viżwali professjonali

Prompt avvanzat jinqasam f’saffi, li kull wieħed iwieġeb mistoqsija distinta: min jew x’inhu, kif rranġat, kif imdawwal, minn fejn qed jitħares, f’liema reġistru, u b’liema karatteristiċi ottiċi. Li tsemmi dawn is-saffi b’mod espliċitu fit-test tal-prompt—“Suġġett: … Kompożizzjoni: … Dawl: …”—jgħin lill-mudell jevita li jħallat ir-reġistri. L-eżempju hawn taħt jiddeskrivi arkitett fl-istudjo tiegħu: is-suġġett jokkupa t-tielet tax-xellug skont ir-regola tat-tliet terzi, id-dawl jiġi lateralment minn twieqi kbar, il-kamera tinsab f’angolu baxx sottili b’tul fokali ta’ 35mm, ir-reġistru huwa fotografija dokumentarja editorjali b’toni sħan desaturati, u l-fond tal-kamp iħalli l-isfond f’bokeh ħafif. Kull waħda minn dawn is-sitt sentenzi tneħħi libertà li l-mudell kieku kieku jeżerċita b’mod każwali.

from openai import OpenAI
import base64

client = OpenAI()

def prompt_multicouche(
    sujet: str,
    composition: str,
    eclairage: str,
    perspective: str,
    style: str,
    details_techniques: str,
    output: str
):
    """Jibni prompt b'diversi saffi għal kontroll massimu."""
    prompt = (
        f"Subject: {sujet}. "
        f"Composition: {composition}. "
        f"Lighting: {eclairage}. "
        f"Perspective: {perspective}. "
        f"Style: {style}. "
        f"Technical details: {details_techniques}."
    )

    response = client.images.generate(
        model="gpt-image-2.5-sunburst",
        prompt=prompt,
        size="1536x1024",
        quality="high",
    )

    data = base64.b64decode(response.data[0].b64_json)
    with open(output, "wb") as f:
        f.write(data)
    print(f"Advanced image: {output}")
    return prompt


prompt_utilise = prompt_multicouche(
    sujet="Architect working on blueprints in a bright studio",
    composition="Rule of thirds, subject at left third, depth with multiple planes",
    eclairage="Lateral natural light from large windows, soft shadows",
    perspective="Slightly low angle (subtle low-angle shot), 35mm focal length",
    style="Editorial documentary photography, warm desaturated tones",
    details_techniques="Medium depth of field, sharp foreground, light background bokeh",
    output="architecte_atelier.png"
)

Tgħallim tal-kompożizzjoni

Il-kompożizzjoni tiddetermina kif l-għajn tivvjaġġa minn ġo immaġni, u int tikkontrollaha billi tsemmi r-regoli li l-mudell jirrikonoxxi.

Regoli klassiċi tal-kompożizzjoni

Id-dizzjunarju hawn taħt jiġbor ħames familji kbar ta’ kompożizzjoni u japplikahom għal suġġett intenzjonalment identiku—fanal iżolat fuq irdum Brittaniku—sabiex il-konfront jiffoka biss fuq l-istadjar. Ir-regola tat-tliet terzi tpoġġi l-fanal f’intersezzjoni ta’ linji ta’ forza u tipproduċi l-aktar immaġni “naturali”; is-simetrija ta’ Wes Anderson tiċċentra l-assi vertikali u żżid riflessjoni fl-ilma; id-dijagonali toħloq tensjoni mill-kantuniera t’isfel tax-xellug sal-kantuniera ta’ fuq tal-lemin; il-qafas ġewwa qafas iġiegħlek tara l-fanal permezz ta’ arkata jew tieqa, u toħloq fond immedjat; fl-aħħar nett, il-kompożizzjoni minimalista tnaqqas il-fanal għal punt żgħir f’ispazju negattiv vast, u tbiddel is-suġġett f’pretest. Ħames immaġnijiet tal-istess fanal, ħames dikjarazzjonijiet differenti.

from openai import OpenAI
import base64

client = OpenAI()

compositions = {
    "regle_des_tiers": (
        "Rule of thirds composition, subject positioned at the intersection "
        "of power lines (right third, bottom third)"
    ),
    "symetrie": (
        "Perfectly symmetrical composition, central vertical axis, "
        "reflections in water, Wes Anderson style"
    ),
    "diagonale": (
        "Dynamic diagonal composition, power line from "
        "bottom-left corner to top-right corner"
    ),
    "cadre_dans_cadre": (
        "Frame within a frame: subject viewed through an arch, "
        "a window, or a natural framing"
    ),
    "minimaliste": (
        "Minimalist composition, huge negative space, "
        "small isolated subject in a vast environment"
    ),
}

sujet_base = "An isolated lighthouse on a Breton cliff facing the sea"

for nom, composition in compositions.items():
    prompt = f"{sujet_base}, {composition}, landscape photography"

    response = client.images.generate(
        model="gpt-image-2.5-sunburst",
        prompt=prompt,
        size="1536x1024",
        quality="high",
    )

    path = f"composition_{nom}.png"
    data = base64.b64decode(response.data[0].b64_json)
    with open(path, "wb") as f:
        f.write(data)
    print(f"Composition {nom} → {path}")

Pjani u fond

Il-fond ma jintalabx bl-ingrossa; jinbena pjan b’pjan. Il-prompt li ġej jiddeskrivi suq Provenzali billi jassenja b’mod espliċitu kontenut u ċarezza lil kull wieħed mit-tliet pjani: sapun u lavanda fl-ewwel pjan, perfettament ċari; il-bejjiegħ u l-klijent tiegħu fil-pjan tan-nofs, kemxejn imċajpar; faċċati tal-ġebel dehbi u siġar tal-platanu fl-isfond, b’bokeh pronunzjat. Is-semmi finali ta’ lenti ta’ 50mm miftuħa għal f/1.8 tikkonferma ottikament biss dak li d-deskrizzjoni diġà stabbilixxiet—u hija din ir-ridondanza intenzjonali li tagħmel ir-riżultat affidabbli.

from openai import OpenAI
import base64

client = OpenAI()

prompt_profondeur = (
    "Provençal market scene with three distinct planes: "
    "sharp foreground—display of lavender and artisanal soaps, "
    "middle ground—vendor and customer in conversation, slight blur, "
    "background—golden stone facades and plane trees, pronounced bokeh. "
    "Photography with 50mm f/1.8 lens, shallow depth of field."
)

response = client.images.generate(
    model="gpt-image-2.5-sunburst",
    prompt=prompt_profondeur,
    size="1536x1024",
    quality="high",
)

data = base64.b64decode(response.data[0].b64_json)
with open("marche_profondeur.png", "wb") as f:
    f.write(data)

Tgħallim tad-dawl

Id-dawl huwa l-aktar għodda qawwija biex toħloq atmosfera, u l-aktar effiċjenti f’termini ta’ karattri investiti. Sitt setups ikopru l-biċċa l-kbira tal-ħtiġijiet, u l-lukkap hawn taħt japplikahom kollha għall-istess ritratt ta’ vjolinista konċentrat. Id-dawl ta’ Rembrandt, bis-sors wieħed tiegħu f’45 grad u l-highlight triangolari fuq il-għadma opposta, jagħti r-ritratt klassiku tal-istudjo. Ir-rim light ipoġġi s-sors wara s-suġġett u jżomm biss l-awreola tal-kontorn, u jipproduċi immaġni ta’ siluwett. Id-dawl maqsum jaqsam il-wiċċ f’żewġ nofsijiet, wieħed imdawwal u l-ieħor fid-dell, u jibdel immedjatament għal reġistru ċinematografiku. Is-siegħa tad-deheb (golden hour) iġib dawl dehbi radenti u l-lens flare nostalġiku tiegħu, in-neon urban iġib sorsi roża u ċjan riflessi fuq wiċċ imxarrab, u l-istudjo high-key iġib uniformità mingħajr dell u sfond abjad kummerċjali. L-istess suġġett isir intimu, misterjuż jew lest għar-reklamar skont il-każ.

from openai import OpenAI
import base64

client = OpenAI()

eclairages_avances = {
    "rembrandt": (
        "Rembrandt lighting: single source at 45 degrees, triangular highlight "
        "on opposite cheek, deep dramatic shadows"
    ),
    "rim_light": (
        "Backlighting (rim light): silhouette outlined by a luminous halo, "
        "source directly behind subject, details in shadows"
    ),
    "split": (
        "Split lighting: half face lit, half in shadow, "
        "clear separation, mysterious cinematic atmosphere"
    ),
    "golden_hour": (
        "Golden hour: raking golden light, long warm shadows, "
        "subtle lens flare, soft nostalgic atmosphere"
    ),
    "neon": (
        "Urban neon lighting: multiple pink and cyan sources, "
        "reflections on wet surfaces, nocturnal cyberpunk atmosphere"
    ),
    "studio_high_key": (
        "High-key studio: uniformly bright lighting, minimal shadows, "
        "white background, clean commercial atmosphere"
    ),
}

sujet = "Portrait of a focused violinist"

for nom, eclairage in eclairages_avances.items():
    prompt = f"{sujet}, {eclairage}"

    response = client.images.generate(
        model="gpt-image-2.5-sunburst",
        prompt=prompt,
        size="1024x1536",
        quality="high",
    )

    path = f"eclairage_{nom}.png"
    data = base64.b64decode(response.data[0].b64_json)
    with open(path, "wb") as f:
        f.write(data)
    print(f"Lighting {nom} → {path}")

Tgħallim tal-perspettiva

Il-perspettiva tinfluwenza kif it-telespettatur jipperċepixxi s-suġġett, u l-effett tagħha huwa kważi dejjem psikoloġiku qabel ma jkun estetiku. Angolu għoli (high angle) iħares lejn ix-xena minn fuq u jnaqqas dak li qed iħares; angolu baxx (low angle), bil-kamera fil-livell tal-art, jikber is-suġġett u jagħtih qawwa; veduta ajrua perfettament vertikali tabolixxi l-volum u tbiddel ir-realtà f’mudell grafiku; veduta tad-dudu (worm’s-eye view) tiċċaqlaq l-angolu baxx sal-estrem, u tagħmel il-linji kollha tal-vanishing point jikkonverġu lejn is-sema; fl-aħħar nett, veduta fil-livell tal-għajn b’tul fokali ta’ 50mm tirriproduċi l-viżjoni umana u tisparixxi favur is-suġġett. Fuq skrapers tal-ħġieġ u l-azzar, id-differenza bejn dawn il-ħames immaġnijiet ma tinsabx fil-bini iżfir-relazzjoni li jistabbilixxu miegħu. Innota li l-lukkap ibiddel għal format portrait għaż-żewġ angoli baxxi, għax shot b’angolu baxx ta’ bini jeħtieġ għoli biex jieħu n-nifs.

from openai import OpenAI
import base64

client = OpenAI()

perspectives = {
    "plongee": (
        "High-angle view (top-down), camera overlooking the scene, "
        "domination effect, subject viewed from above at 45 degrees"
    ),
    "contre_plongee": (
        "Low-angle view (bottom-up), camera at ground level looking up, "
        "effect of grandeur and power"
    ),
    "aerienne": (
        "Drone aerial view, perfectly vertical (90 degrees), "
        "graphic composition, textures and patterns visible from above"
    ),
    "worm_eye": (
        "Worm's-eye view, camera placed on ground, extreme perspective "
        "with vanishing lines converging toward the sky"
    ),
    "eye_level": (
        "Eye-level view, natural and immersive, "
        "50mm focal length, like human vision"
    ),
}

sujet = "Modern glass and steel skyscrapers"

for nom, perspective in perspectives.items():
    prompt = f"{sujet}, {perspective}, architectural photography"

    response = client.images.generate(
        model="gpt-image-2.5-sunburst",
        prompt=prompt,
        size="1024x1536" if nom in ["contre_plongee", "worm_eye"] else "1536x1024",
        quality="high",
    )

    path = f"perspective_{nom}.png"
    data = base64.b64decode(response.data[0].b64_json)
    with open(path, "wb") as f:
        f.write(data)
    print(f"Perspective {nom} → {path}")

Builder tal-prompt riutilizzabbli

Li tgħaqqad seba’ saffi bl-idejn f’kull script qabel jew wara se jġiegħlek tinsa wieħed jew taqleb żewġ oħrajn. Il-klassi PromptBuilder issolvi dan billi tesponi metodu għal kull saff u tirritorna self f’kull sejħa, u b’hekk tippermettilek tgħaqqad is-settings f’ordni li tinqara u tispeċifika biss uħud minnhom. Il-metodu build() imbagħad jgħaqqad is-suġġett u s-saffi mimlija f’test lest biex jgħaddih lil client.images.generate(). L-eżempju tal-kafè fuq ix-xatt jimplimenta s-seba’ saffi disponibbli kollha—kompożizzjoni, dawl, perspettiva, stil, kuluri, atmosfera, teknika—u jipproduċi prompt li l-ebda improvviżazzjoni ma kienet tagħmel daqshekk sħiħ.

class PromptBuilder:
    """Builder għal prompts viżwali strutturati."""

    def __init__(self, sujet: str):
        self.sujet = sujet
        self.layers = {}

    def composition(self, desc: str):
        self.layers["composition"] = desc
        return self

    def eclairage(self, desc: str):
        self.layers["lighting"] = desc
        return self

    def perspective(self, desc: str):
        self.layers["perspective"] = desc
        return self

    def style(self, desc: str):
        self.layers["style"] = desc
        return self

    def couleurs(self, desc: str):
        self.layers["color palette"] = desc
        return self

    def atmosphere(self, desc: str):
        self.layers["atmosphere"] = desc
        return self

    def technique(self, desc: str):
        self.layers["technique"] = desc
        return self

    def build(self) -> str:
        parts = [self.sujet]
        for key, value in self.layers.items():
            parts.append(f"{key}: {value}")
        return ". ".join(parts)


# Użu
prompt = (
    PromptBuilder("Artisanal seaside café, terrace with umbrellas")
    .composition("Rule of thirds, terrace in foreground, sea at horizon")
    .eclairage("Late afternoon, raking golden light, long shadows")
    .perspective("Slight high angle, 28mm wide-angle focal length")
    .style("Lifestyle photography, warm tones, analog film")
    .couleurs("Terracotta, Mediterranean blue, off-white, olive green")
    .atmosphere("Relaxed and Mediterranean, sea breeze")
    .technique("Deep depth of field, sharp from foreground to infinity")
    .build()
)

print(prompt)
# Uża dan il-prompt ma' client.images.generate()

Issa ħu pussess ta’ din il-kaxxa tal-għodda fuq suġġett wieħed li jitlob ħafna: librerija antika. Ipproduċi ħames immaġnijiet bl-użu ta’ PromptBuilder, u varja sistematikament il-kompożizzjoni, id-dawl u l-perspettiva, imbagħad iġborhom f’katalgu HTML. Li tara l-ħames immaġnijiet ħdejn xulxin huwa kif se titgħallem liema minn dawn is-saffi jimpurtaw l-aktar fil-prattika tiegħek.