ruff

app.py

@@ -8,7 +8,6 @@ import numpy as np
 import torch
 from gradio.themes import Soft
 from PIL import Image, ImageDraw, ImageFont
-
 from transformers import Sam3TrackerVideoModel, Sam3TrackerVideoProcessor, Sam3VideoModel, Sam3VideoProcessor
 
 
@@ -153,7 +152,7 @@ def init_video_session(
     device = _GLOBAL_DEVICE
     dtype = _GLOBAL_DTYPE
 
-    video_path: Optional[str] = None
+    video_path: str | None = None
     if isinstance(video, dict):
         video_path = video.get("name") or video.get("path") or video.get("data")
     elif isinstance(video, str):
@@ -286,7 +285,7 @@ def compose_frame(state: AppState, frame_idx: int) -> Image.Image:
                 try:
                     font = ImageFont.truetype(font_path, font_size)
                     break
-                except (OSError, IOError):
+                except OSError:
                     continue
             if font is None:
                 # Fallback to default font
@@ -338,7 +337,7 @@ def update_frame_display(state: AppState, frame_idx: int) -> Image.Image:
     return compose_frame(state, frame_idx)
 
 
-def _get_prompt_for_obj(state: AppState, obj_id: int) -> Optional[str]:
+def _get_prompt_for_obj(state: AppState, obj_id: int) -> str | None:
     """Get the prompt text associated with an object ID."""
     # Priority 1: Check text_prompts_by_frame_obj (most reliable)
     for frame_texts in state.text_prompts_by_frame_obj.values():
@@ -415,29 +414,28 @@ def on_image_click(
             state.pending_box_start_obj_id = ann_obj_id
             state.composited_frames.pop(ann_frame_idx, None)
             return update_frame_display(state, ann_frame_idx)
-        else:
-            x1, y1 = state.pending_box_start
-            x2, y2 = int(x), int(y)
-            state.pending_box_start = None
-            state.pending_box_start_frame_idx = None
-            state.pending_box_start_obj_id = None
-            state.composited_frames.pop(ann_frame_idx, None)
-            x_min, y_min = min(x1, x2), min(y1, y2)
-            x_max, y_max = max(x1, x2), max(y1, y2)
-
-            box = [[[x_min, y_min, x_max, y_max]]]
-            processor.add_inputs_to_inference_session(
-                inference_session=state.inference_session,
-                frame_idx=ann_frame_idx,
-                obj_ids=ann_obj_id,
-                input_boxes=box,
-            )
+        x1, y1 = state.pending_box_start
+        x2, y2 = int(x), int(y)
+        state.pending_box_start = None
+        state.pending_box_start_frame_idx = None
+        state.pending_box_start_obj_id = None
+        state.composited_frames.pop(ann_frame_idx, None)
+        x_min, y_min = min(x1, x2), min(y1, y2)
+        x_max, y_max = max(x1, x2), max(y1, y2)
 
-            frame_boxes = state.boxes_by_frame_obj.setdefault(ann_frame_idx, {})
-            obj_boxes = frame_boxes.setdefault(ann_obj_id, [])
-            obj_boxes.clear()
-            obj_boxes.append((x_min, y_min, x_max, y_max))
-            state.composited_frames.pop(ann_frame_idx, None)
+        box = [[[x_min, y_min, x_max, y_max]]]
+        processor.add_inputs_to_inference_session(
+            inference_session=state.inference_session,
+            frame_idx=ann_frame_idx,
+            obj_ids=ann_obj_id,
+            input_boxes=box,
+        )
+
+        frame_boxes = state.boxes_by_frame_obj.setdefault(ann_frame_idx, {})
+        obj_boxes = frame_boxes.setdefault(ann_obj_id, [])
+        obj_boxes.clear()
+        obj_boxes.append((x_min, y_min, x_max, y_max))
+        state.composited_frames.pop(ann_frame_idx, None)
     else:
         label_int = 1 if str(label).lower().startswith("pos") else 0
 
@@ -654,7 +652,7 @@ def propagate_masks(GLOBAL_STATE: gr.State):
                 return
 
             # Add all prompts to the inference session (processor handles deduplication)
-            for text_prompt in text_prompt_to_obj_ids.keys():
+            for text_prompt in text_prompt_to_obj_ids:
                 GLOBAL_STATE.inference_session = processor.add_text_prompt(
                     inference_session=GLOBAL_STATE.inference_session,
                     text=text_prompt,
@@ -840,17 +838,16 @@ def reset_session(GLOBAL_STATE: gr.State) -> tuple[AppState, Image.Image, int, i
         GLOBAL_STATE.inference_session, "reset_inference_session"
     ):
         GLOBAL_STATE.inference_session.reset_inference_session()
-    else:
-        if GLOBAL_STATE.video_frames:
-            processor = _GLOBAL_TRACKER_PROCESSOR
-            raw_video = [np.array(frame) for frame in GLOBAL_STATE.video_frames]
-            GLOBAL_STATE.inference_session = processor.init_video_session(
-                video=raw_video,
-                inference_device=_GLOBAL_DEVICE,
-                video_storage_device="cpu",
-                processing_device="cpu",
-                dtype=_GLOBAL_DTYPE,
-            )
+    elif GLOBAL_STATE.video_frames:
+        processor = _GLOBAL_TRACKER_PROCESSOR
+        raw_video = [np.array(frame) for frame in GLOBAL_STATE.video_frames]
+        GLOBAL_STATE.inference_session = processor.init_video_session(
+            video=raw_video,
+            inference_device=_GLOBAL_DEVICE,
+            video_storage_device="cpu",
+            processing_device="cpu",
+            dtype=_GLOBAL_DTYPE,
+        )
 
     GLOBAL_STATE.masks_by_frame.clear()
     GLOBAL_STATE.clicks_by_frame_obj.clear()

pyproject.toml

@@ -0,0 +1,60 @@
+[project]
+name = "sam3-video-segmentation"
+version = "0.1.0"
+description = "Add your description here"
+readme = "README.md"
+requires-python = ">=3.10"
+dependencies = [
+    "accelerate>=1.11.0",
+    "gradio>=5.49.1",
+    "imageio[pyav]>=2.37.2",
+    "kernels>=0.11.0",
+    "opencv-python>=4.12.0.88",
+    "spaces>=0.42.1",
+    "torch==2.8.0",
+    "torchvision>=0.23.0",
+    "transformers",
+]
+
+[tool.ruff]
+line-length = 119
+
+[tool.ruff.lint]
+select = ["ALL"]
+ignore = [
+    "COM812", # missing-trailing-comma
+    "D203",   # one-blank-line-before-class
+    "D213",   # multi-line-summary-second-line
+    "E501",   # line-too-long
+    "SIM117", # multiple-with-statements
+    #
+    "D100",    # undocumented-public-module
+    "D101",    # undocumented-public-class
+    "D102",    # undocumented-public-method
+    "D103",    # undocumented-public-function
+    "D104",    # undocumented-public-package
+    "D105",    # undocumented-magic-method
+    "D107",    # undocumented-public-init
+    "EM101",   # raw-string-in-exception
+    "FBT001",  # boolean-type-hint-positional-argument
+    "FBT002",  # boolean-default-value-positional-argument
+    "PGH003",  # blanket-type-ignore
+    "PLR0913", # too-many-arguments
+    "PLR0915", # too-many-statements
+    "TRY003",  # raise-vanilla-args
+]
+unfixable = [
+    "F401", # unused-import
+]
+
+[tool.ruff.lint.pydocstyle]
+convention = "google"
+
+[tool.ruff.lint.per-file-ignores]
+"*.ipynb" = ["T201", "T203"]
+
+[tool.ruff.format]
+docstring-code-format = true
+
+[tool.uv.sources]
+transformers = { git = "https://github.com/huggingface/transformers.git", rev = "69f003696b" }