Suya OCR API — vLLM-backed, OpenAI-compatible OCR service

FastAPI service wrapping the Surya-OCR-2 model (datalab-to) served through vLLM: legacy /v1/api/ai/* endpoints, an OpenAI-compatible /v1/chat/completions endpoint, a coalescing request batcher, a local OCR CLI, Docker packaging, multilingual example outputs, and quantization/concurrency benchmarks. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-17 10:20:02 +04:00
commit 1a585693be
147 changed files with 13827 additions and 0 deletions
@@ -0,0 +1,44 @@
+from typing import List
+
+from surya.common.polygon import PolygonBox
+
+
+def clean_boxes(boxes: List[PolygonBox]) -> List[PolygonBox]:
+    new_boxes = []
+    for box_obj in boxes:
+        xs = [point[0] for point in box_obj.polygon]
+        ys = [point[1] for point in box_obj.polygon]
+        if max(xs) == min(xs) or max(ys) == min(ys):
+            continue
+
+        box = box_obj.bbox
+        contained = False
+        for other_box_obj in boxes:
+            if other_box_obj.polygon == box_obj.polygon:
+                continue
+
+            other_box = other_box_obj.bbox
+            if box == other_box:
+                continue
+            if (
+                box[0] >= other_box[0]
+                and box[1] >= other_box[1]
+                and box[2] <= other_box[2]
+                and box[3] <= other_box[3]
+            ):
+                contained = True
+                break
+        if not contained:
+            new_boxes.append(box_obj)
+    return new_boxes
+
+
+def expand_bbox(bbox, expansion_factor=0.01):
+    expansion_low = 1 - expansion_factor
+    expansion_high = 1 + expansion_factor
+    return [
+        bbox[0] * expansion_low,
+        bbox[1] * expansion_low,
+        bbox[2] * expansion_high,
+        bbox[3] * expansion_high,
+    ]