# javierfigueroa11/detectorpersonas architecture

> Architecture diagram of the GitHub repository javierfigueroa11/detectorpersonas, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-11-19). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/javierfigueroa11/detectorpersonas
- Repository: https://github.com/javierfigueroa11/detectorpersonas

## Overview

DetectorPersonas is a Python video-analysis application: a user selects YOLO, a fine-tuned YOLO model, or Grounding DINO, then the app processes video frames, counts detections, and writes annotated video and CSV results. A separate SAM 2 entry point segments an object across a supplied frame directory and produces a video. The README is minimal; connections below are limited to the sampled source, so training and evaluation artifacts are not mapped as runtime stages.

Main entry points: [`main.py`](https://github.com/javierfigueroa11/detectorpersonas/blob/main/main.py) (Application entry), [`clases/procesador_videos.py`](https://github.com/javierfigueroa11/detectorpersonas/blob/main/clases/procesador_videos.py) (Video processor), [`clases/segmentador_sam2.py`](https://github.com/javierfigueroa11/detectorpersonas/blob/main/clases/segmentador_sam2.py) (SAM 2 segmenter), [`clases/contador_personas.py`](https://github.com/javierfigueroa11/detectorpersonas/blob/main/clases/contador_personas.py) (Person counter)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_workflow["Video workflow"]
  node_main["Application entry<br/>[main.py]"]
  node_source_videos["Input videos"]
  node_processor["Video processor"]
  node_counter["Person counter"]
end

subgraph group_detection["Detection models"]
  node_yolo["YOLO detector"]
  node_grounding["Grounding DINO"]
  node_grounding_hub["Model repository"]
  node_finetuned_weights["Fine-tuned weights"]
end

subgraph group_tracking["Segmentation"]
  node_sam2["SAM 2 segmenter"]
  node_frames["Input frames"]
end

subgraph group_outputs["Results"]
  node_video_results["Video outputs"]
  node_csv_results["CSV outputs"]
end

node_user(("User"))

node_user -->|"selects mode"| node_main
node_main -->|"reads videos"| node_source_videos
node_main -->|"selects detector"| node_yolo
node_main -->|"selects detector"| node_grounding
node_main -->|"loads weights"| node_finetuned_weights
node_main -->|"processes video"| node_processor
node_processor -->|"detects frames"| node_yolo
node_processor -->|"detects and annotates"| node_grounding
node_processor -->|"records detections"| node_counter
node_processor -->|"writes annotated video"| node_video_results
node_main -->|"saves CSV"| node_counter
node_counter -->|"writes CSV"| node_csv_results
node_grounding -->|"loads pretrained model"| node_grounding_hub
node_user -->|"requests segmentation"| node_main
node_main -->|"runs segmentation"| node_sam2
node_main -->|"passes frame directory"| node_frames
node_sam2 -->|"reads and propagates"| node_frames
node_sam2 -->|"writes segmented video"| node_video_results

click node_main "https://github.com/javierfigueroa11/detectorpersonas/blob/main/main.py"
click node_source_videos "https://github.com/javierfigueroa11/detectorpersonas/tree/main/data_videos"
click node_processor "https://github.com/javierfigueroa11/detectorpersonas/blob/main/clases/procesador_videos.py"
click node_yolo "https://github.com/javierfigueroa11/detectorpersonas/blob/main/clases/detector_personas.py"
click node_grounding "https://github.com/javierfigueroa11/detectorpersonas/blob/main/clases/detector_grounding_dino.py"
click node_counter "https://github.com/javierfigueroa11/detectorpersonas/blob/main/clases/contador_personas.py"
click node_sam2 "https://github.com/javierfigueroa11/detectorpersonas/blob/main/clases/segmentador_sam2.py"
click node_frames "https://github.com/javierfigueroa11/detectorpersonas/tree/main/dataset_personas/frames_extraidos"
click node_video_results "https://github.com/javierfigueroa11/detectorpersonas/tree/main/resultados"
click node_csv_results "https://github.com/javierfigueroa11/detectorpersonas/tree/main/resultados"
click node_finetuned_weights "https://github.com/javierfigueroa11/detectorpersonas/blob/main/modelos_entrenados/yolov8n_finetuning_v2.pt"

classDef toneNeutral fill:#f8fafc,stroke:#334155,stroke-width:1.5px,color:#0f172a
classDef toneBlue fill:#dbeafe,stroke:#2563eb,stroke-width:1.5px,color:#172554
classDef toneAmber fill:#fef3c7,stroke:#d97706,stroke-width:1.5px,color:#78350f
classDef toneMint fill:#dcfce7,stroke:#16a34a,stroke-width:1.5px,color:#14532d
classDef toneRose fill:#ffe4e6,stroke:#e11d48,stroke-width:1.5px,color:#881337
classDef toneIndigo fill:#e0e7ff,stroke:#4f46e5,stroke-width:1.5px,color:#312e81
classDef toneTeal fill:#ccfbf1,stroke:#0f766e,stroke-width:1.5px,color:#134e4a
class node_main,node_source_videos,node_processor,node_counter,node_user toneBlue
class node_yolo,node_grounding,node_grounding_hub,node_finetuned_weights toneAmber
class node_sam2,node_frames toneMint
class node_video_results,node_csv_results toneRose
```

## Components

### Video workflow

- [Application entry](https://github.com/javierfigueroa11/detectorpersonas/blob/main/main.py): `main.py`
- [Input videos](https://github.com/javierfigueroa11/detectorpersonas/tree/main/data_videos): `data_videos`
- [Video processor](https://github.com/javierfigueroa11/detectorpersonas/blob/main/clases/procesador_videos.py): `clases/procesador_videos.py`
- [Person counter](https://github.com/javierfigueroa11/detectorpersonas/blob/main/clases/contador_personas.py): `clases/contador_personas.py`

### Detection models

- [YOLO detector](https://github.com/javierfigueroa11/detectorpersonas/blob/main/clases/detector_personas.py): `clases/detector_personas.py`
- [Grounding DINO](https://github.com/javierfigueroa11/detectorpersonas/blob/main/clases/detector_grounding_dino.py): `clases/detector_grounding_dino.py`
- **Model repository**
- [Fine-tuned weights](https://github.com/javierfigueroa11/detectorpersonas/blob/main/modelos_entrenados/yolov8n_finetuning_v2.pt): `modelos_entrenados/yolov8n_finetuning_v2.pt`

### Segmentation

- [SAM 2 segmenter](https://github.com/javierfigueroa11/detectorpersonas/blob/main/clases/segmentador_sam2.py): `clases/segmentador_sam2.py`
- [Input frames](https://github.com/javierfigueroa11/detectorpersonas/tree/main/dataset_personas/frames_extraidos): `dataset_personas/frames_extraidos`

### Results

- [Video outputs](https://github.com/javierfigueroa11/detectorpersonas/tree/main/resultados): `resultados`
- [CSV outputs](https://github.com/javierfigueroa11/detectorpersonas/tree/main/resultados): `resultados`

### Other components

- **User**

## Connections

- User → Application entry: selects mode (evidence: `main.py`)
- Application entry → Input videos: reads videos (evidence: `main.py`)
- Application entry → YOLO detector: selects detector (evidence: `main.py`)
- Application entry → Grounding DINO: selects detector (evidence: `main.py`)
- Application entry → Fine-tuned weights: loads weights (evidence: `main.py`)
- Application entry → Video processor: processes video (evidence: `main.py`)
- Video processor → YOLO detector: detects frames (evidence: `clases/procesador_videos.py`)
- Video processor → Grounding DINO: detects and annotates (evidence: `clases/procesador_videos.py`)
- Video processor → Person counter: records detections (evidence: `clases/procesador_videos.py`)
- Video processor → Video outputs: writes annotated video (evidence: `clases/procesador_videos.py`)
- Application entry → Person counter: saves CSV (evidence: `main.py`)
- Person counter → CSV outputs: writes CSV (evidence: `clases/contador_personas.py`)
- Grounding DINO → Model repository: loads pretrained model (evidence: `clases/detector_grounding_dino.py`)
- User → Application entry: requests segmentation (evidence: `main.py`)
- Application entry → SAM 2 segmenter: runs segmentation (evidence: `main.py`)
- Application entry → Input frames: passes frame directory (evidence: `main.py`)
- SAM 2 segmenter → Input frames: reads and propagates (evidence: `clases/segmentador_sam2.py`)
- SAM 2 segmenter → Video outputs: writes segmented video (evidence: `clases/segmentador_sam2.py`)
