Drop dead detect_background_color; sync docs with landed optimizations

- chroma.py: remove detect_background_color — orphaned since auto-detection
  goes through estimate_background_model -> _background_border_cluster;
  nothing in the repo calls it.
- settings.py: fix stale "Segmentation always stays fp32" comment
  (--precision now fans out to the BiRefNet models too).
- README: add a measured CPU performance section (9700X reference numbers,
  memory ceiling explanation, Zen 2 fallback guidance).
- docs/hair_gap_artifacts.md: record that the cross-check cost note is
  obsolete — reuse_as_seg returns the net model count to 2, bf16 and
  chunked attention absorb the rest.
- docs/green_screen_matting_workflow.md: add the 2026-07 additions to the
  architecture-evolution note.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

README.md

@@ -114,6 +114,26 @@ D:\MiniConda\envs\lightML\python.exe -m bgfilter.cli `
   --debug-dir Outputs\debug
 ```
 
+## Performance (CPU reference numbers)
+
+Measured on a Ryzen 9700X (Zen 5, native bf16), 32 GB RAM, default settings
+(cross-check on, reused as seg; chunked global attention on):
+
+| input | precision | warm / image | peak memory |
+|---|---|---|---|
+| 1024x1536 | fp32 | ~45 s | — |
+| 1024x1536 | bf16 | ~33 s | — |
+| 2048x2048 | bf16 | ~53 s | 11.4 GB (batch), 8.1 GB (single image) |
+
+The memory ceiling is the cross-check BiRefNet_HR forward (fixed
+`input_size` 2048 regardless of the image size) — genuine live activations
+of a full-resolution dense prediction net. Before chunked attention and the
+mimalloc fix, the same 2048x2048 bf16 run peaked at 25.2 GB; a fully
+un-optimized fp32 run would need an estimated 55-60 GB (ViTMatte's un-chunked
+N^2 attention alone ~39 GB). CPUs without native bf16 (e.g. Zen 2) auto-fall
+back to fp32 — on such machines prefer `--no-cross-check` if memory or
+latency is tight.
+
 ## Chroma-alpha debug mode
 
 `--matting-method chroma` skips ViTMatte and uses chroma confidence directly as the

bgfilter/chroma.py

@@ -112,22 +112,6 @@ def _background_border_cluster(
     return full, share, agree
 
 
-def detect_background_color(
-    rgb: np.ndarray, settings: ChromaSettings | None = None
-) -> tuple[float, float, float]:
-    """Detect a single flat background colour from the image border.
-
-    Returns the colour as a 0..1 RGB triple. Raises RuntimeError (via
-    :func:`_background_border_cluster`) if the border is not dominated by one flat
-    colour, so callers fail loudly rather than matting against a wrong colour.
-    Pure: reads pixels only, with no matting side effects.
-    """
-    settings = settings or ChromaSettings()
-    rgb_f, lab = convert_color_spaces(rgb)
-    mask, _, _ = _background_border_cluster(lab, settings)
-    return tuple(float(x) for x in np.median(rgb_f[mask], axis=0))
-
-
 def estimate_background_model(
     rgb: np.ndarray,
     settings: ChromaSettings,

bgfilter/settings.py

@@ -129,10 +129,10 @@ class ModelSettings:
     matting_method: str = "vitmatte"
     fallback_to_chroma_alpha: bool = False
     # ViTMatte compute precision: "fp32" or "bf16". bf16 halves the matting
-    # model's activation memory (its full-resolution attention is the pipeline's
-    # memory peak) with visually identical alpha, but needs bf16-capable
-    # hardware (any modern GPU, or a CPU with AVX512-BF16/AMX) — otherwise it
-    # falls back to fp32 with a warning. Segmentation always stays fp32.
+    # model's activation memory with visually identical alpha, but needs
+    # bf16-capable hardware (any modern GPU, or a CPU with AVX512-BF16/AMX) —
+    # otherwise it falls back to fp32 with a warning. The BiRefNet models have
+    # their own precision fields; --precision sets all of them at once.
     precision: str = "fp32"
     # Query-chunked global attention (exact math, bitwise-identical output):
     # the VitDet backbone's 4 global blocks materialize an N^2 attention map —

docs/green_screen_matting_workflow.md

@@ -11,6 +11,7 @@
 - **两条管线**（`segmentation.enabled` 切换）：无分割模型的 Chroma + ViTMatte；或单个语义分割模型（默认 birefnet，可切 anime-seg，用 `--seg-backend`）定 trimap 拓扑，ViTMatte 细化。
 - **背景色泛化**：`screen_color` 先验（如 `#CFEFFF`）；`null` 时自动从图像边框探测平背景色（探测不到干净平背景会报错），不再写死绿幕。
 - **已移除**：染绿修复（recolor）、发丝补救（dropout-fill）、BiRefNet+anime-seg 结合方案——它们只是给“生图即染色/断裂”的坏素材擦屁股；源头改用干净背景生图后不再需要。
+- **新增（2026-07）**：交叉模型否决（`cross_check`，BiRefNet_HR-matting 第二意见 min 融合，清发丝缝隙残蓝，见 `docs/hair_gap_artifacts.md`）；其前向默认兼任 seg mask（`reuse_as_seg`，净模型数回到 2）；`--precision bf16` 一个旋钮加速全部三类模型；ViTMatte 全局 attention 默认按 query 分块（输出逐位不变，2048² 内存峰值 25→11 GB）。
 
 下文其余章节描述绿幕基线流程，仍然有效。
 

docs/hair_gap_artifacts.md

@@ -89,6 +89,14 @@ min 融合;B2 的意见锐化(remap 0.5→0.95)被否——数字更低(0.673 vs
 **已落地(2026-07-04,默认开启)。** 新增 `cross_check` 设置块 +
 `--cross-check/--no-cross-check`;pipeline 在 clean_alpha 后插入
 `cross_check_alpha`(alpha_post.py),第二意见加载复用 BiRefNetSegmenter。
+
+**成本已消化(2026-07-06)。** 当年"+1 个模型 +1 次推理"的账已改写:
+`cross_check.reuse_as_seg`(默认开)让 HR-matting 前向兼任 seg mask,
+**净模型数回到 2 个**(trimap 99.8% 不变,先导 Outputs/segreuse/);
+`--precision bf16` 经 autocast + deform_conv2d shim 覆盖 BiRefNet
+(35s→26s @2048);ViTMatte 全局 attention 分块(逐位等价)把内存峰值
+从 25.2GB 压到 11.4GB(2048×2048 bf16,9700X)。CPU 上开 cross-check
+的常规图约 33s/张(bf16)。
 验证:开启时 TestImage3 输出与评审原型逐字节一致;`--no-cross-check` 与
 旧 baseline 逐字节一致;FixImage1(粉底)脸部无损、发缘粉残留被清,
 唯一副作用是羽化权重外溢到暗豁免区边缘(弓弦零星像素最多 -0.5,中位