kinemotion 0.39.1__tar.gz → 0.41.0__tar.gz

kinemotion-0.41.0/.basic-memory/development/comprehensive-timing-instrumentation.md

@@ -0,0 +1,274 @@
+---
+title: Comprehensive Timing Instrumentation
+type: note
+permalink: development/comprehensive-timing-instrumentation
+---
+
+# Comprehensive Timing Instrumentation
+
+## Overview
+
+Complete timing instrumentation implemented across the entire backend and core analysis pipeline. Every significant operation now emits structured timing events for full end-to-end visibility. Analysis reveals **MediaPipe pose tracking dominates processing time (88-94%)**.
+
+## Architecture & Data Flow
+
+### Complete Data Flow
+```
+process_cmj_video / process_dropjump_video (API)
+    ↓
+    stage_times dict (collected during analysis)
+    ├─ timing_video_initialization
+    ├─ timing_pose_tracking
+    ├─ timing_parameter_auto_tuning
+    ├─ timing_smoothing
+    ├─ timing_vertical_position_extraction
+    ├─ timing_phase_detection
+    ├─ timing_metrics_calculation
+    └─ timing_quality_assessment
+    ↓
+ProcessingInfo.timing_breakdown (stored in metrics metadata)
+    ↓
+metrics.to_dict() includes timing_breakdown_ms
+    ↓
+Backend extracts and emits individual timing events
+    ↓
+Structured logs (JSON in production, colored in development)
+```
+
+## Complete List of All Timing Events (20 Total)
+
+### Backend Operations (5 events)
+1. **`timing_video_validation`** - Validating uploaded video file format/codec
+2. **`timing_video_file_save`** - Writing uploaded video file to temporary disk
+3. **`timing_r2_input_video_upload`** - Uploading input video to R2 storage
+4. **`timing_response_serialization`** - Converting metrics to JSON and building HTTP response
+5. **`timing_temp_file_cleanup`** - Cleaning up temporary video files
+
+### Core Analysis Pipeline (8 events)
+6. **`timing_video_initialization`** - VideoProcessor setup and frame reading
+7. **`timing_pose_tracking`** - MediaPipe pose detection on all frames (~90% of total)
+8. **`timing_parameter_auto_tuning`** - Auto-tuning algorithm parameters
+9. **`timing_smoothing`** - Savitzky-Golay smoothing filter
+10. **`timing_vertical_position_extraction`** - Extracting Y-axis positions
+11. **`timing_phase_detection`** - Detecting jump phases
+12. **`timing_metrics_calculation`** - Computing final metrics
+13. **`timing_quality_assessment`** - Quality assessment
+
+### Output Generation (3 events)
+14. **`timing_metadata_building`** - Creating VideoInfo, ProcessingInfo, AlgorithmConfig, ResultMetadata
+15. **`timing_metrics_validation`** - Running DropJumpMetricsValidator or CMJMetricsValidator
+16. **`timing_json_serialization`** - Converting metrics.to_dict() to JSON string
+
+### Debug Video Processing (2 events)
+17. **`timing_debug_video_generation`** - Rendering annotated frames and writing to disk (always logged when requested)
+18. **`timing_debug_video_reencode`** - FFmpeg re-encoding for browser compatibility (only if codec fallback)
+
+### Cloud Storage (2 events)
+19. **`timing_r2_results_upload`** - Uploading metrics JSON to R2
+20. **`timing_r2_debug_video_upload`** - Uploading debug video to R2
+
+## Implementation Details
+
+### Core API Changes (src/kinemotion/api.py)
+
+**Renamed function for consistency:**
+- `_generate_outputs()` → `_generate_dropjump_outputs()`
+- Ensures both jump types use consistent naming pattern
+
+**Updated `_generate_dropjump_outputs()` and `_generate_cmj_outputs()`:**
+- Now returns `tuple[float, float, float]` instead of `None`
+- Returns: `(generation_duration, reencode_duration, json_duration)`
+- Measures JSON serialization, debug video generation, and re-encoding separately
+
+**Updated both `process_dropjump_video()` and `process_cmj_video()`:**
+- Capture timing for metadata building (VideoInfo, ProcessingInfo, AlgorithmConfig, ResultMetadata)
+- Capture timing for metrics validation
+- Unpack all three duration values from output generator
+- Add all timings to `stage_times` dict
+
+### Backend Changes (backend/src/kinemotion_backend/app.py)
+
+**Added timing event emissions:**
+- `timing_video_validation` (line 507) - Validates uploaded video
+- `timing_video_file_save` (line 519) - Writes to temporary disk
+- `timing_r2_input_video_upload` (line 534) - Renamed from `video_uploaded_to_r2` for consistency
+- `timing_response_serialization` (line 654) - JSON serialization overhead
+- `timing_temp_file_cleanup` (line 732) - Temporary file cleanup
+
+All backend timings logged at INFO level with `logger.info()` and `duration_ms` field.
+
+### Metadata Changes (src/kinemotion/core/metadata.py)
+
+**Added to `ProcessingInfo` dataclass:**
+- `timing_breakdown: dict[str, float] | None = None`
+- `to_dict()` converts timings to milliseconds as `timing_breakdown_ms`
+- Included in returned metrics metadata
+
+### Debug Overlay Changes (src/kinemotion/core/debug_overlay_utils.py)
+
+**Updated `BaseDebugOverlayRenderer`:**
+- Added `reencode_duration_s = 0.0` attribute
+- `close()` method captures FFmpeg re-encoding time
+- Only populated if codec fallback occurs (mp4v → H.264)
+
+## Log Output Examples
+
+### Development Mode (Colored)
+```
+2024-12-10T14:23:45.123456+00:00 [INFO] timing_pose_tracking duration_ms=5183.1
+2024-12-10T14:23:45.234567+00:00 [INFO] timing_debug_video_generation duration_ms=850.3
+2024-12-10T14:23:48.234567+00:00 [INFO] timing_debug_video_reencode duration_ms=400.2
+2024-12-10T14:23:49.345678+00:00 [INFO] timing_r2_results_upload duration_ms=234.0
+```
+
+### Production Mode (JSON)
+```json
+{"timestamp": "2025-12-10T14:23:45.123456Z", "severity": "INFO", "event": "timing_pose_tracking", "duration_ms": 5183.1}
+{"timestamp": "2025-12-10T14:23:45.234567Z", "severity": "INFO", "event": "timing_debug_video_generation", "duration_ms": 850.3}
+{"timestamp": "2025-12-10T14:23:48.234567Z", "severity": "INFO", "event": "timing_debug_video_reencode", "duration_ms": 400.2}
+{"timestamp": "2025-12-10T14:23:49.345678Z", "severity": "INFO", "event": "timing_r2_results_upload", "duration_ms": 234.0}
+```
+
+### Metrics JSON Response
+```json
+{
+  "metadata": {
+    "processing": {
+      "version": "0.39.1",
+      "timestamp": "2025-12-10T13:41:33.547498+00:00",
+      "quality_preset": "balanced",
+      "processing_time_s": 5.713,
+      "timing_breakdown_ms": {
+        "Video initialization": 325.4,
+        "Pose tracking": 5183.1,
+        "Parameter auto-tuning": 2.0,
+        "Smoothing": 188.6,
+        "Vertical position extraction": 4.6,
+        "Phase detection": 1.2,
+        "Metrics calculation": 0.2,
+        "Quality assessment": 7.7
+      }
+    }
+  }
+}
+```
+
+## Performance Analysis
+
+### CMJ Analysis (236 frames @ 29.58fps, 8s video)
+```
+Video initialization.....................    128ms (  2.3%)
+Pose tracking...........................   5196ms ( 94.1%)  ← BOTTLENECK
+Parameter auto-tuning...................      2ms (  0.0%)
+Smoothing...............................    179ms (  3.2%)
+Vertical position extraction............      5ms (  0.1%)
+Phase detection.........................      1ms (  0.0%)
+Metrics calculation.....................      0ms (  0.0%)
+Quality assessment......................      8ms (  0.2%)
+Total...................................   5521ms (100.0%)
+```
+
+### Drop Jump Analysis (89 frames @ 29.73fps, 3s video)
+```
+Pose tracking...........................   2114ms ( 92.1%)  ← BOTTLENECK
+Parameter auto-tuning...................      1ms (  0.0%)
+Smoothing...............................     70ms (  3.0%)
+Vertical position extraction............      1ms (  0.0%)
+Ground contact detection................      0ms (  0.0%)
+Metrics calculation.....................      4ms (  0.2%)
+Quality assessment......................      3ms (  0.1%)
+Total...................................   2294ms (100.0%)
+```
+
+### Key Findings
+- **Primary Bottleneck**: MediaPipe pose tracking (88-94% of time)
+  - Processes every frame through neural network
+  - ~22-23ms per frame on average
+  - Scales linearly with frame count
+
+- **Fast Operations**:
+  - Smoothing: 3%
+  - Video initialization: 2-4%
+  - All other operations: <1% each
+
+## Google Cloud Logging Queries
+
+**Find all timing events:**
+```
+jsonPayload.event=~"^timing_"
+```
+
+**Find specific stage:**
+```
+jsonPayload.event="timing_pose_tracking"
+```
+
+**Find R2 operations:**
+```
+jsonPayload.event=~"^timing_r2_"
+```
+
+**Find analysis pipeline only (exclude R2/debug):**
+```
+jsonPayload.event=~"^timing_" AND NOT jsonPayload.event=~"^timing_r2_|timing_debug"
+```
+
+**Find slow stages (>1 second):**
+```
+jsonPayload.event=~"^timing_" AND jsonPayload.duration_ms > 1000
+```
+
+**Timeline of all operations:**
+```
+jsonPayload.event=~"^timing_" AND timestamp>="2025-12-10T14:23:45Z" AND timestamp<"2025-12-10T14:24:00Z"
+```
+
+**Average duration per stage:**
+```
+# Use Cloud Monitoring to aggregate by event name and calculate mean(duration_ms)
+```
+
+## Optimization Recommendations
+
+### High Impact (Address Pose Tracking Bottleneck)
+1. **Reduce frame rate** - Process every Nth frame (e.g., 15fps instead of 30fps)
+2. **GPU acceleration** - MediaPipe supports GPU if available
+3. **Async processing** - Pre-process poses while waiting for upload
+4. **Frame batching** - Process multiple frames in parallel if memory permits
+
+### Medium Impact
+1. **Optimize smoothing** - Already efficient (3%), but FFT possible for large videos
+2. **Lazy output generation** - Only generate debug video if requested
+
+### Low Impact (Already fast)
+- Phase detection, metrics calculation, quality assessment already negligible
+
+## Benefits of Complete Instrumentation
+
+✅ **Full Pipeline Visibility** - See every stage from upload to response
+✅ **Bottleneck Identification** - Pose tracking consistently ~90% (MediaPipe limitation)
+✅ **Backend vs Analysis** - Measure server overhead vs. analysis time
+✅ **Debug Overhead** - Track cost of generating debug videos
+✅ **Storage Costs** - Monitor R2 upload times
+✅ **Response Latency** - Measure serialization overhead
+✅ **Validation Performance** - Track metrics validation timing
+✅ **Metadata Building** - Measure config object creation time
+✅ **Granular Querying** - Each event independently queryable
+✅ **Optimization Targets** - Data-driven decisions on improvements
+
+## Testing
+
+✅ All 38 API tests pass (cmj and dropjump)
+✅ All 85 backend tests pass
+✅ Type checking passes (pyright strict)
+✅ Linting passes (ruff)
+✅ Timing data flows correctly through entire pipeline
+✅ JSON output includes timing_breakdown_ms
+✅ Individual timing events properly logged at INFO level
+
+## Files Modified
+
+- `src/kinemotion/api.py` - Renamed `_generate_outputs()`, added timing capture for all stages
+- `src/kinemotion/core/metadata.py` - Added `timing_breakdown` field to `ProcessingInfo`
+- `src/kinemotion/core/debug_overlay_utils.py` - Added `reencode_duration_s` attribute
+- `backend/src/kinemotion_backend/app.py` - Added 5 new timing event emissions

{kinemotion-0.39.1 → kinemotion-0.41.0}/CHANGELOG.md

@@ -7,6 +7,35 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 <!-- version list -->
 
+## v0.41.0 (2025-12-10)
+
+### Features
+
+- Add comprehensive timing instrumentation across backend and core pipeline
+  ([`acb3df6`](https://github.com/feniix/kinemotion/commit/acb3df65d2895fbdae7fcc7f52a73bf7b6fb36d1))
+
+- Implement hybrid timing instrumentation with PerformanceTimer abstraction
+  ([`b207193`](https://github.com/feniix/kinemotion/commit/b207193e047c869e52dfe5303616806f491f938f))
+
+
+## v0.40.0 (2025-12-10)
+
+### Documentation
+
+- Switch formatting tool from black to ruff in GEMINI.md
+  ([`7bf2e3e`](https://github.com/feniix/kinemotion/commit/7bf2e3e073fea368470359a0bdff65b8568e48e1))
+
+### Features
+
+- Add detailed timing logs to analysis pipeline
+  ([`4c75820`](https://github.com/feniix/kinemotion/commit/4c758201be03dbd06f463e2434ca287bb123bfd5))
+
+### Testing
+
+- Relax flight time assertion in kinematics test
+  ([`b360c4d`](https://github.com/feniix/kinemotion/commit/b360c4de88c92e365657775041fe7eea04892a27))
+
+
 ## v0.39.1 (2025-12-10)
 
 ### Bug Fixes

{kinemotion-0.39.1 → kinemotion-0.41.0}/GEMINI.md

@@ -20,18 +20,18 @@ The project uses `uv` for dependency management and `asdf` for Python version ma
 **Key Libraries:**
 
 - **Production**: `click`, `opencv-python`, `mediapipe`, `numpy`, `scipy`.
-- **Development**: `pytest`, `black`, `ruff`, `mypy`.
+- **Development**: `pytest`, `ruff`, `pyright`.
 
 ### Development Commands
 
 - **Run CLI**: `uv run kinemotion dropjump-analyze <video_path>`
 - **Install/Sync Dependencies**: `uv sync`
 - **Run Tests**: `uv run pytest`
-- **Format Code**: `uv run black src/`
+- **Format Code**: `uv run ruff format .`
 - **Lint Code**: `uv run ruff check`
 - **Auto-fix Linting**: `uv run ruff check --fix`
-- **Type Check**: `uv run mypy src/kinemotion`
-- **Run All Checks**: `uv run ruff check && uv run mypy src/kinemotion && uv run pytest`
+- **Type Check**: `uv run pyright`
+- **Run All Checks**: `uv run ruff check && uv run pyright && uv run pytest`
 
 ## Architecture
 
@@ -127,18 +127,20 @@ docs/                   # Documentation (PARAMETERS.md is key)
 
 ## Code Quality & Workflow
 
+**CRITICAL**: Never commit code that fails sanity checks. Always run `ruff`, `pyright`, and `pytest` locally before committing.
+
 When contributing code, strictly adhere to the project's quality standards.
 
-1. **Format Code**: `uv run black src/`
+1. **Format Code**: `uv run ruff format .`
 1. **Lint and Fix**: `uv run ruff check --fix`
-1. **Type Check**: `uv run mypy src/kinemotion`
+1. **Type Check**: `uv run pyright`
 1. **Run Tests**: `uv run pytest`
 
-**Run all checks before committing**: `uv run ruff check && uv run mypy src/kinemotion && uv run pytest`
+**Run all checks before committing**: `uv run ruff check && uv run pyright && uv run pytest`
 
-- **Type Safety**: The project uses `mypy` in strict mode. All functions must have full type annotations.
+- **Type Safety**: The project uses `pyright` in strict mode. All functions must have full type annotations.
 - **Linting**: `ruff` is used for linting. Configuration is in `pyproject.toml`.
-- **Formatting**: `black` is used for code formatting.
+- **Formatting**: `ruff` is used for code formatting.
 
 ## Common Development Tasks
 

{kinemotion-0.39.1 → kinemotion-0.41.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: kinemotion
-Version: 0.39.1
+Version: 0.41.0
 Summary: Video-based kinematic analysis for athletic performance
 Project-URL: Homepage, https://github.com/feniix/kinemotion
 Project-URL: Repository, https://github.com/feniix/kinemotion

{kinemotion-0.39.1 → kinemotion-0.41.0}/backend/src/kinemotion_backend/app.py

@@ -24,6 +24,7 @@ from fastapi import (
 from fastapi.middleware.cors import CORSMiddleware
 from fastapi.responses import JSONResponse
 from kinemotion.api import process_cmj_video, process_dropjump_video
+from kinemotion.core.timing import PerformanceTimer
 from slowapi import Limiter, _rate_limit_exceeded_handler
 from slowapi.util import get_remote_address
 
@@ -395,6 +396,7 @@ async def _process_video_async(
     jump_type: JumpType,
     quality: str = "balanced",
     output_video: str | None = None,
+    timer: PerformanceTimer | None = None,
 ) -> dict[str, Any]:
     """Process video and return metrics.
 
@@ -403,6 +405,7 @@ async def _process_video_async(
         jump_type: Type of jump analysis
         quality: Analysis quality preset
         output_video: Optional path for debug video output
+        timer: Optional PerformanceTimer for measuring operations
 
     Returns:
         Dictionary with metrics
@@ -412,11 +415,11 @@ async def _process_video_async(
     """
     if jump_type == "drop_jump":
         metrics = process_dropjump_video(
-            video_path, quality=quality, output_video=output_video
+            video_path, quality=quality, output_video=output_video, timer=timer
         )
     else:  # cmj
         metrics = process_cmj_video(
-            video_path, quality=quality, output_video=output_video
+            video_path, quality=quality, output_video=output_video, timer=timer
         )
 
     return cast(dict[str, Any], metrics.to_dict())
@@ -499,15 +502,26 @@ async def analyze_video(
         # Normalize to lowercase
         jump_type = cast(JumpType, _validate_jump_type(jump_type))
         await file.seek(0)
+
+        # Validate video file
+        validation_start = time.time()
         _validate_video_file(file)
+        validation_duration = time.time() - validation_start
+        logger.info(
+            "timing_video_validation",
+            duration_ms=round(validation_duration * 1000),
+        )
 
         # Create temporary file for processing
         suffix = Path(file.filename or "video.mp4").suffix
         with tempfile.NamedTemporaryFile(suffix=suffix, delete=False) as temp_file:
             temp_video_path = temp_file.name
             # Write uploaded file to temp location
+            save_start = time.time()
             content = await file.read()
             temp_file.write(content)
+            save_duration = time.time() - save_start
+        logger.info("timing_video_file_save", duration_ms=round(save_duration * 1000))
 
         # Create temporary path for debug video output
         with tempfile.NamedTemporaryFile(suffix=".mp4", delete=False) as temp_debug:
@@ -515,10 +529,16 @@ async def analyze_video(
 
         # Upload to R2 if client available
         if r2_client:
+            upload_start = time.time()
             r2_video_key = f"videos/{jump_type}/{file.filename}"
             try:
                 r2_client.upload_file(temp_video_path, r2_video_key)
-                logger.info("video_uploaded_to_r2", key=r2_video_key)
+                upload_duration = time.time() - upload_start
+                logger.info(
+                    "timing_r2_input_video_upload",
+                    key=r2_video_key,
+                    duration_ms=round(upload_duration * 1000),
+                )
             except OSError as e:
                 logger.error("r2_upload_failed", error=str(e), key=r2_video_key)
                 raise HTTPException(
@@ -527,17 +547,41 @@ async def analyze_video(
                 ) from e
 
         # Process video with real kinemotion analysis
+        analysis_start = time.time()
+        timer = PerformanceTimer()
         metrics = await _process_video_async(
             temp_video_path,
             jump_type,
             quality,
             output_video=temp_debug_video_path,
+            timer=timer,
         )  # type: ignore[arg-type]
+        analysis_duration = time.time() - analysis_start
+
+        # Log detailed timing breakdown if available in metadata
+        if (
+            "metadata" in metrics
+            and "processing" in metrics["metadata"]
+            and "timing_breakdown_ms" in metrics["metadata"]["processing"]
+        ):
+            timing_breakdown = metrics["metadata"]["processing"]["timing_breakdown_ms"]
+            # Normalize keys for easier jq parsing: spaces → underscores, lowercase
+            normalized_timings = {
+                stage.lower().replace(" ", "_") + "_ms": duration
+                for stage, duration in timing_breakdown.items()
+            }
+            # Log each timing stage as a separate event for granular monitoring
+            for stage_key, duration_ms in normalized_timings.items():
+                # Remove "_ms" suffix and create timing event name
+                stage_name = stage_key.replace("_ms", "")
+                event_name = f"timing_{stage_name}"
+                logger.info(event_name, duration_ms=round(duration_ms, 1))
 
         logger.info(
             "video_analysis_completed",
             jump_type=jump_type,
-            metrics_count=len(metrics),
+            metrics_count=len(metrics.get("data", {})),
+            duration_ms=round(analysis_duration * 1000),
         )
 
         # Upload results and debug video to R2 if client available
@@ -552,12 +596,17 @@ async def analyze_video(
             # Upload Metrics JSON
             r2_results_key = f"results/{jump_type}/{file_stem}_results.json"
             try:
+                results_upload_start = time.time()
                 results_json = json.dumps(metrics, indent=2)
                 results_url = r2_client.put_object(
                     r2_results_key, results_json.encode()
                 )
+                results_upload_duration = time.time() - results_upload_start
                 logger.info(
-                    "results_uploaded_to_r2", key=r2_results_key, url=results_url
+                    "timing_r2_results_upload",
+                    key=r2_results_key,
+                    url=results_url,
+                    duration_ms=round(results_upload_duration * 1000),
                 )
             except OSError as e:
                 # Log error but don't fail - results still available in response
@@ -573,13 +622,16 @@ async def analyze_video(
             ):
                 r2_debug_video_key = f"debug_videos/{jump_type}/{file_stem}_debug.mp4"
                 try:
+                    debug_upload_start = time.time()
                     debug_video_url = r2_client.upload_file(
                         temp_debug_video_path, r2_debug_video_key
                     )
+                    debug_upload_duration = time.time() - debug_upload_start
                     logger.info(
-                        "debug_video_uploaded_to_r2",
+                        "timing_r2_debug_video_upload",
                         key=r2_debug_video_key,
                         url=debug_video_url,
+                        duration_ms=round(debug_upload_duration * 1000),
                     )
                 except OSError as e:
                     logger.warning(
@@ -600,9 +652,18 @@ async def analyze_video(
             processing_time_s=processing_time,
         )
 
+        # Serialize response to JSON
+        serialization_start = time.time()
+        response_dict = response.to_dict()
+        serialization_duration = time.time() - serialization_start
+        logger.info(
+            "timing_response_serialization",
+            duration_ms=round(serialization_duration * 1000),
+        )
+
         return JSONResponse(
             status_code=200,
-            content=response.to_dict(),
+            content=response_dict,
         )
 
     except ValueError as e:
@@ -646,6 +707,9 @@ async def analyze_video(
         )
 
     finally:
+        # Clean up temporary files
+        cleanup_start = time.time()
+
         # Clean up temporary video file
         if temp_video_path and Path(temp_video_path).exists():
             try:
@@ -668,6 +732,13 @@ async def analyze_video(
                     error=str(e),
                 )
 
+        cleanup_duration = time.time() - cleanup_start
+        if cleanup_duration > 0:
+            logger.info(
+                "timing_temp_file_cleanup",
+                duration_ms=round(cleanup_duration * 1000),
+            )
+
         # Clean up R2 video if results failed (optional - adjust based on policy)
         # if r2_client and r2_video_key and not results_url:
         #     try:
@@ -705,7 +776,7 @@ async def analyze_local_video(
 
     try:
         # Validate inputs
-        jump_type = _validate_jump_type(jump_type)  # Normalize to lowercase
+        jump_type = cast(JumpType, _validate_jump_type(jump_type))
 
         if not Path(video_path).exists():
             raise HTTPException(
@@ -714,7 +785,10 @@ async def analyze_local_video(
             )
 
         # Process video
-        metrics = await _process_video_async(video_path, jump_type, quality)  # type: ignore[arg-type]
+        timer = PerformanceTimer()
+        metrics = await _process_video_async(
+            video_path, jump_type, quality, timer=timer
+        )  # type: ignore[arg-type]
 
         processing_time = time.time() - start_time
 

{kinemotion-0.39.1 → kinemotion-0.41.0}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "kinemotion"
-version = "0.39.1"
+version = "0.41.0"
 description = "Video-based kinematic analysis for athletic performance"
 readme = "README.md"
 requires-python = ">=3.10,<3.13"

{kinemotion-0.39.1 → kinemotion-0.41.0}/sonar-project.properties

@@ -2,7 +2,7 @@
 sonar.projectKey=feniix_kinemotion
 sonar.projectName=Kinemotion
 sonar.organization=feniix
-sonar.projectVersion=0.35.0
+sonar.projectVersion=0.38.0
 
 # Project metadata
 sonar.projectDescription=Video-based kinematic analysis for athletic performance using MediaPipe pose tracking