我的妹妹其實沒有那麼可愛 五更琉璃 photographing practice 02
第一張黑白 | 第二張彩色 | |
氣氛 | 很強 | 強 |
角色辨識 | 較弱 | 很強 |
光影造型 | 最好 | 很好 |
公仔材質感 | 被弱化 | 比較明顯 |
底座存在感 | 很低 | 稍強 |
人像攝影感 | 很強 | 中等偏強 |
商品展示感 | 弱 | 較強 |
這兩張其實很有意思，因為構圖與燈位幾乎相同，但最後呈現的是兩種不同的作品方向。如果把它當成「公仔商品照」與「低調人像作品」之間來看，我反而覺得你已經比較偏後者了。
第一張：黑白
我比較喜歡第一張的整體氣氛。
黑白把公仔原本「塑膠模型」的資訊抽掉很多，觀眾注意力會直接落在臉部輪廓、肩膀、手臂、腰線到腿部的曲線。左上方來的窄光也因此變得非常重要，尤其臉頰 → 肩膀 → 手臂這條亮部線條很漂亮。
而且你留了大量左側負空間，人物又朝左看，視線前方有空間，所以畫面不是單純「把公仔放在右邊」，而有一種人物正在看向黑暗中的感覺。這點我認為是這張最成功的地方。
黑白版還有一個優勢：底座消失得比較自然。第一眼幾乎不是在看「公仔＋底座」，而比較像在看一張棚拍低調人像。
不過它目前稍微暗了一點。特別是頭髮後半部、腰部、裙子與右手附近已經接近完全融入背景。如果再暗一點，人物的腰部會有「斷掉」的感覺。
我不會把整張曝光提高，反而會局部處理：臉部維持現在亮度，肩膀與手臂可以不動，只把頭髮右側輪廓、腰部／裙擺提高約 0.2～0.4 EV。這樣仍然是 low-key，但人物形體會完整很多。
第二張：彩色
第二張的優點恰好相反：角色資訊比較完整。
眼睛的紅色、頭飾紫色、膚色、黑白服裝，以及底座偏紫的色彩都出來了，所以比較容易看出這是一尊精緻的動漫公仔。
而且臉上的微弱暖色很漂亮，沒有一般公仔照片那種「塑膠皮膚被打得很亮」的問題。你的光源控制得相當克制。
但彩色版有一個我比較在意的地方：
紫色底座開始搶戲了。
因為整個畫面大約 90% 都是黑、深灰、低彩度，最下面突然出現一大片紫色，人眼很容易被拉到下面。尤其底座又是水平橢圓形，視覺重量很大。
如果是我後製，我會把底座：
- 飽和度降低一些
- 明度再壓約 0.3 EV
- 紫色不要完全消掉，只留下淡淡的色彩暗示
這樣眼睛的紅色和頭飾的紫色反而會成為真正的視覺亮點。
兩張放在一起，我會這樣看
所以不是單純哪張「比較漂亮」，而是兩張在往不同方向走。
如果你想拍的是「公仔攝影作品」，我會保留第一張。
因為第一張已經有一點脫離「我拍了一尊公仔」的感覺，更接近用公仔作為人像攝影的被攝體。這其實跟你之前說用 Platypod、G2 等器材拍公仔的方向很吻合：你現在值得繼續發展的可能不是傳統均勻打光的模型照，而是這種 low-key portrait。
還有一個我很喜歡的小地方：**兩張都沒有把臉照滿。**鼻子、嘴巴、下巴留在陰影交界附近，使角色有立體感。如果正面再補一盞柔光把臉照清楚，我反而覺得會破壞現在的氣氛。
如果讓我再拍一次，我會保持主燈位置幾乎不變，只在人物右後方非常弱地加一道 rim light（輪廓光），大概比主燈低 2.5～3.5 EV，只讓黑髮、背部和裙子邊緣從背景分離出一條極細的線。
那樣第一張黑白版應該會非常漂亮。
不過呢....Cheney 覺得還是有很大的進步空間....這只是其中的兩張
比較能看的
大部分的照片....顏色都很沉.....感覺顏色都糊在一起了
就像下面的這張
所以又針對這樣的問題 Cheney 問了ChatGPT
以下是他的回應
對，這張我很明顯感覺到你說的「沉重」。而且跟上一張相比，這張更容易看出問題在哪裡。
不是顏色不夠鮮艷，而是整張照片的「明暗配置」讓顏色顯得厚重。
你看這張幾個特徵：黑背景占很大比例；頭髮是接近黑色的深藍；衣服也是深藍黑；陰影又壓得很深。真正明亮的區域主要只剩皮膚和白色蕾絲。因此整體視覺自然會變成「黑＋深藍＋膚色」，紅眼和紫色髮飾雖然有彩度，但面積非常小。
更關鍵的是你的主光偏硬，而且明暗反差很大。你看鼻子下面、瀏海下面、脖子、右手臂內側、裙子右側，都有很深的陰影。這種光其實很適合戲劇性的 low-key portrait，但它天然就不會有「清爽、鮮艷」的感覺。
我反而不建議你先調 Saturation
如果直接把這張 Saturation +20、+30，紅眼和紫色髮飾會變得更紅更紫，但整張照片還是沉重。
我會先改「光」。
你下一次保持這個黑背景、主燈位置、相機曝光都不變，在相機附近、公仔正前方偏右放你前面提到的白色壓克力板。
大概可以想成：
主燈
↘
↘
[公仔]
白色反光板 📷
↗
補暗部
反光板不要一開始靠很近。先大約 30～50 cm，拍一張；再移到 20～30 cm 拍一張比較。
你真正要觀察的不是「臉有沒有變亮」，而是：
瀏海下面有沒有開始出現細節、脖子的黑影有沒有柔和、深藍衣服能不能看出藍色、右側手臂是不是不再掉進黑裡。
只要這四個地方開始出現資訊，整張照片就會突然從「沉」變成比較「通透」。
還有一個很重要的地方：背景不用換
我甚至建議你繼續用黑背景。
因為黑背景＋彩色公仔本來可以非常鮮艷。真正漂亮的做法不是把背景換成白色，而是做到：
背景維持 0～5% 的亮度，人物暗部提高到大約 15～25%，皮膚中間調再提高一些，而高光不要爆。
這樣黑背景仍然很黑，但是人物不會跟著黑。
你這張其實已經很接近了。尤其左眼的紅、紫色髮飾、膚色都證明色彩本身沒有問題。現在缺的是暗部的光，而不是更多的色。
另外，如果你想保留現在這種很有戲劇性的主燈，我會優先用白色反光板，而不是第二盞燈。反光板補出來的暗部通常更自然，也比較不會把你現在漂亮的立體感破壞掉。
如果你的目標是「黑背景，但顏色看起來清澈、鮮艷，不沉重」，我覺得這會是你下一次拍攝最值得做的一個改變。
至於Cheney根據ChatGPT的建議後的照片成果,就放在下一篇囉
敬請期待
Continuing from the previous post, Cheney finally managed to get a couple of shots that looked pretty decent. And, as usual, he threw them over to ChatGPT to see what it had to say.
First, let's start with a table.
(AI really is useful... In the old days, I would've had to make the table myself, figure out all the categories, organize everything... Now I just hand it to AI and it's done.)
| Category | First Photo — B&W | Second Photo — Color |
|---|---|---|
| Atmosphere | Very strong | Strong |
| Character recognition | Weaker | Very strong |
| Light & shadow shaping | Best | Very good |
| Figure material/texture | Reduced | More apparent |
| Base visibility | Very low | Slightly stronger |
| Portrait photography feel | Very strong | Moderately strong |
| Product photography feel | Weak | Stronger |
Now let's take a look at ChatGPT's comments.
These two images are actually very interesting because the composition and lighting positions are almost identical, yet the final results head in two completely different directions.
If we think of them as sitting somewhere between "figure product photography" and "low-key portrait photography," I actually think you've already moved much closer to the latter.
First Photo: Black & White
I prefer the overall atmosphere of the first image.
Converting it to black and white removes a lot of the visual information that reminds us we're looking at a "plastic figure." Instead, the viewer's attention goes directly to the contours of the face, shoulder, arm, waistline, and the curves extending down toward the legs.
Because of this, the narrow light coming from the upper left becomes especially important. The highlighted line running from the cheek → shoulder → arm is particularly beautiful.
You've also left a large amount of negative space on the left side of the frame, while the character is looking toward the left. That gives her gaze somewhere to go.
As a result, the composition doesn't simply feel like "a figure placed on the right side of the frame." Instead, it creates the impression that the character is staring into the darkness.
I think that's the most successful aspect of this image.
The black-and-white version has another advantage: the base disappears much more naturally.
At first glance, you almost don't see it as "a figure + a display base." It feels more like a low-key studio portrait.
That said, the image is currently just a little too dark.
In particular, the back of the hair, the waist, the skirt, and the area around the right hand are getting very close to disappearing completely into the background. If those areas were any darker, the character's body would start to feel visually disconnected around the waist.
I wouldn't increase the exposure of the entire image.
Instead, I'd make localized adjustments: keep the face at its current brightness, leave the shoulder and arm alone, and raise only the right-side outline of the hair and the waist/skirt area by roughly 0.2–0.4 EV.
That would preserve the low-key atmosphere while making the character's form feel much more complete.
Second Photo: Color
The second image has almost the opposite advantage: it preserves much more information about the character.
The red eyes, purple hair accessory, skin tone, black-and-white outfit, and the slightly purple color of the base are all visible. That makes it much easier to recognize this as a finely detailed anime figure.
The subtle warmth on the face also looks very nice. It avoids the common problem in figure photography where the lighting makes the skin look overly bright and plasticky.
Your lighting is actually quite restrained here.
There is, however, one thing that bothers me more in the color version:
The purple base is beginning to steal attention from the character.
Roughly 90% of the image consists of black, dark gray, and low-saturation colors. Then suddenly there's a relatively large area of purple at the bottom of the frame, so the viewer's eye is naturally pulled downward.
The base is also a horizontal oval, which gives it considerable visual weight.
If I were editing this image, I would:
Reduce the saturation of the base slightly
Lower its brightness by around 0.3 EV
Keep a hint of purple rather than removing the color completely
That way, the red eyes and purple hair accessory would become the true color accents in the image.
Looking at the Two Images Together
So I don't think this is simply a question of which image is "prettier."
They're moving in two different directions.
If your goal is to create a "figure photography artwork," I would keep the first image.
The first shot is already beginning to move away from the feeling of "I photographed a figure."
Instead, it feels more like using a figure as the subject of portrait photography.
This actually fits very well with the direction you've been exploring with equipment such as the Platypod and G2. What may be worth developing further isn't traditional, evenly lit figure photography, but this kind of low-key portrait photography.
There's also one small detail I really like:
Neither image fully illuminates the face.
The nose, mouth, and chin remain close to the boundary between light and shadow, which gives the character a stronger sense of three-dimensional form.
If you added another soft light from the front and illuminated the entire face clearly, I actually think it would weaken the atmosphere you've created.
If I were shooting this again, I'd leave the main light almost exactly where it is and add only an extremely subtle rim light behind and to the right of the character.
I'd set it roughly 2.5–3.5 EV below the main light—just enough to create a very thin line separating the black hair, back, and edge of the skirt from the background.
I think the black-and-white version would look especially beautiful with that adjustment.
But...
Cheney still feels there's a lot of room for improvement.
These were only two of the better-looking shots.
Most of the photos...
The colors looked very heavy and muddy.
Everything seemed to blend together.
Something like the photo below.
So Cheney went back to ChatGPT again and asked about this problem.
Here's what it said:
Yes, I can clearly see what you mean by the image feeling "heavy" in this shot. Compared with the previous image, it's actually easier to identify what's causing the problem here.
The problem isn't that the colors aren't saturated enough. It's the overall distribution of light and shadow that's making the colors feel dense and heavy.
Look at a few characteristics of this image.
The black background occupies a large portion of the frame. The hair is an extremely dark blue that's almost black. The outfit is also a very dark blue-black, and the shadows are pushed quite deep.
The only truly bright areas left are mainly the skin and the white lace.
As a result, the overall visual impression naturally becomes:
black + dark blue + skin tone.
The red eyes and purple hair accessory do have saturation, but they occupy only a very small portion of the image.
More importantly, your main light is relatively hard, and the contrast between light and shadow is quite strong.
Look at the shadows beneath the nose, under the bangs, around the neck, on the inside of the right arm, and along the right side of the skirt.
They're all quite deep.
This type of lighting works extremely well for dramatic low-key portraits, but by nature it won't produce a "bright, clear, colorful" appearance.
I Actually Wouldn't Adjust Saturation First
If you simply increase Saturation by +20 or +30, the red eyes will become redder and the purple hair accessory more purple—but the image will still feel heavy.
I'd change the lighting first.
For your next shoot, keep the same black background, main-light position, and camera exposure.
Then place the white acrylic board you mentioned earlier near the camera, slightly to the front-right of the figure, and use it as a reflector.
Think of the setup roughly like this:
Main Light
↘
↘
[Figure]
White Reflector 📷
↗
Fill ShadowsDon't place the reflector too close at first.
Start at around 30–50 cm, take a shot, then move it to around 20–30 cm and take another shot for comparison.
What you really want to observe isn't simply whether "the face becomes brighter."
Instead, look for these things:
Are details beginning to appear beneath the bangs?
Are the deep shadows around the neck becoming softer?
Can you actually see that the dark-blue clothing is blue?
Is the right arm still disappearing into black?
As soon as those four areas begin to reveal more information, the entire image will suddenly shift from feeling "heavy" to feeling much clearer and more open.
And There's Another Important Point: Don't Change the Background
In fact, I would recommend that you continue using the black background.
A black background combined with a colorful figure can look extremely vivid.
The key isn't to replace the background with white.
Instead, the goal is something like this:
Keep the background at around 0–5% brightness, raise the shadow areas of the figure to roughly 15–25%, lift the skin midtones slightly, and keep the highlights from clipping.
That way, the background stays black without dragging the figure into darkness with it.
This shot is actually already quite close.
The red of the left eye, the purple hair accessory, and the skin tone all show that the colors themselves aren't the problem.
What's missing is light in the shadows—not more color.
Also, if you want to preserve the dramatic quality of your current main light, I would prioritize using a white reflector rather than adding a second light.
Fill created by a reflector usually looks more natural and is less likely to destroy the beautiful three-dimensional modeling you've already created.
If your goal is:
"A black background, but with colors that look clean, vivid, and transparent rather than dark and heavy,"
then I think this is the single most worthwhile change to try during your next shoot.
So, what did Cheney's photos look like after following ChatGPT's advice?
Well...
That's for the next post.
Stay tuned!
Comments