This is a continuation of a previous article1 on BBAVectors and Transformer-based context aware detection:

1. Sample for BBAVectors:

import os

import torch

from PIL import Image

from torchvision import transforms

from models.detector import build_detector # from BBAVectors repo

from utils.visualize import visualize_detections # optional visualization

from utils.inference import run_inference # custom helper you may need to define

# Load pretrained BBAVectors model

def load_bbavectors_model(config_path, checkpoint_path):

    model = build_detector(config_path)

    model.load_state_dict(torch.load(checkpoint_path, map_location='cpu'))

    model.eval()

    return model

# Preprocess image from URI

def load_image_from_uri(uri):

    image = Image.open(uri).convert("RGB")

    transform = transforms.Compose([

        transforms.Resize((1024, 1024)),

        transforms.ToTensor(),

    ])

    return transform(image).unsqueeze(0) # Add batch dimension

# Run detection

def detect_landmarks(model, image_tensor):

    with torch.no_grad():

        outputs = model(image_tensor)

    return outputs # BBAVectors returns oriented bounding boxes

# Main workflow

def main():

    # Paths to config and weights

    config_path = 'configs/dota_bbavectors.yaml'

    checkpoint_path = 'checkpoints/bbavectors_dota.pth'

    # URIs to drone images

    image_uris = [

        'drone_images/scene1.jpg',

        'drone_images/scene2.jpg'

    ]

    model = load_bbavectors_model(config_path, checkpoint_path)

    for uri in image_uris:

        image_tensor = load_image_from_uri(uri)

        detections = detect_landmarks(model, image_tensor)

        print(f"\nDetections for {uri}:")

        for det in detections:

            print(f"Class: {det['label']}, Score: {det['score']:.2f}, BBox: {det['bbox']}")

        # Optional: visualize results

        # visualize_detections(uri, detections)

if __name__ == "__main__":

    main()

2. Sample for semantic based detection:

from PIL import Image

import requests

import torch

from transformers import DetrImageProcessor, DetrForObjectDetection

# Load pretrained DETR model and processor

processor = DetrImageProcessor.from_pretrained("facebook/detr-resnet-50")

model = DetrForObjectDetection.from_pretrained("facebook/detr-resnet-50")

# Function to load image from URI

def load_image(uri):

    return Image.open(requests.get(uri, stream=True).raw).convert("RGB")

# Function to detect objects and return labels

def detect_objects(image):

    inputs = processor(images=image, return_tensors="pt")

    outputs = model(**inputs)

    # Filter predictions by confidence threshold

    target_sizes = torch.tensor([image.size[::-1]])

    results = processor.post_process_object_detection(outputs, target_sizes=target_sizes, threshold=0.9)[0]

    labels = [model.config.id2label[label.item()] for label in results["labels"]]

    return set(labels)

# URIs for two drone-captured scenes

scene1_uri = "https://example.com/drone_scene_1.jpg"

scene2_uri = "https://example.com/drone_scene_2.jpg"

# Load and process both scenes

scene1 = load_image(scene1_uri)

scene2 = load_image(scene2_uri)

labels1 = detect_objects(scene1)

labels2 = detect_objects(scene2)

# Compare object presence

shared_objects = labels1.intersection(labels2)

unique_to_scene1 = labels1 - labels2

unique_to_scene2 = labels2 - labels1

# Print results

print("Shared objects between scenes:", shared_objects)

print("Unique to Scene 1:", unique_to_scene1)

print("Unique to Scene 2:", unique_to_scene2)

 Scene/Object correlation in Aerial Drone Image Analysis

Given aerial drone images and their vector representations for scene and objects, correlation at scene-level and object-level is evolving research in drone sensing applications. The ability to predict object presence in unseen urban scenes using vector representations makes several drone sensing use cases easy to implement on the analytics side without requiring custom models1. Two promising approaches—Box Boundary-Aware Vectors (BBAVectors) and Context-Aware Detection via Transformer and CLIP tokens—offer distinct yet complementary pathways toward this goal. Both methods seek to bridge the semantic gap between scene-level embeddings and object-level features, enabling predictive inference across spatial domains. These are described in the following sections.

Box Boundary-Aware Vectors: Geometry as a Signature

BBAVectors reimagine object detection by encoding geometric relationships rather than relying solely on bounding box regression. Traditional object detectors predict the coordinates of bounding boxes directly, which can be brittle in aerial imagery where objects are rotated, occluded, or densely packed. BBAVectors instead regress directional vectors—top, right, bottom, and left—from the object center to its boundaries. This vectorized representation captures the shape, orientation, and spatial extent of objects in a way that is more robust to rotation and scale variance.

In the context of scene-object correlation, BBAVectors serve as a geometric signature. For example, consider a building with a circular roof in an aerial image2. Its BBAVector profile—equal-length vectors radiating symmetrically from the center—would differ markedly from that of a rectangular warehouse or a triangular-roofed church. When applied to a new scene, the presence of similar BBAVector patterns can suggest the existence of a circular-roofed structure, even if the building is partially occluded or viewed from a different angle.

This approach has been validated in datasets like DOTA (Dataset for Object Detection in Aerial Images)3, where BBAVector-based models outperform traditional detectors in identifying rotated and irregularly shaped objects. By embedding these vectors into a shared latent space, one can correlate object-level geometry with scene-level context, enabling predictive modeling across scenes.

Context-Aware Detection via Transformer and CLIP Tokens: Semantics and Attention

While BBAVectors excel at capturing geometry, context-aware detection leverages semantic relationships. This method treats object proposals and image segments as tokens in a Transformer architecture, allowing the model to learn inter-object and object-background dependencies through attention mechanisms. By integrating CLIP (Contrastive Language–Image Pretraining) features, the model embeds both visual and textual semantics into a unified space.

CLIP tokens encode high-level concepts—such as “circular building,” “parking lot,” or “green space”—based on large-scale image-text training. When combined with Transformer attention, the model can infer the likelihood of object presence based on surrounding context. For instance, if a circular-roofed building is typically adjacent to a park and a road intersection, the model can learn this spatial-semantic pattern. In a new scene with similar context vectors, it can predict the probable presence of the landmark even if it’s not directly visible.

This approach has been explored in works like “DETR” (DEtection TRansformer)4 and “GLIP” (Grounded Language-Image Pretraining)5, which demonstrate how attention-based models can generalize object detection across domains. In aerial imagery, this means that scene-level embeddings—augmented with CLIP tokens—can serve as priors for object-level inference.

Bridging the Two: Predictive Correlation Across Scenes

Together, BBAVectors and context-aware detection offer a dual lens: one geometric, the other semantic. By embedding both object-level vectors and scene-level features into a shared space—whether through contrastive learning, metric learning, or attention-weighted fusion—researchers can build models that predict object presence in new scenes with remarkable accuracy.

Imagine a workflow where a drone captures a new urban scene. The scene is encoded using CLIP-based features and Transformer attention maps. Simultaneously, known object signatures from previous scenes—represented as BBAVectors—are matched against the new scene’s embeddings. If the context and geometry align, the model flags the likely presence of a circular-roofed building, even before it’s explicitly detected.

This paradigm has implications for smart city planning, disaster response, and autonomous navigation. By correlating scene and object vectors, systems can anticipate infrastructure layouts, identify critical assets, and adapt to dynamic environments—all from the air.

#Codingexercise: CodingExercise-10-11-2025.docx 

 Scale Resolution 

Analytics also help with distance and area calculation from aerial drone images from disparate points of time and real-world GPS coordinates.  
The following prompts and responses explain these: 

Case 1: 

Prompt: 

Given the attached aerial drone image and its Estimated Coordinates: Latitude = 42.3736, Longitude = -71.1097 and its scale resolution as 2cm in image=50feet in actual distance, find the square footage occupied by the building with a circular structure. 

Response: 

The square footage occupied by the building with the circular structure in the attached aerial image is approximately 81,300 square feet.frame5.jpg 

Calculation Method 

• The circular structure was located and its size was estimated using the image's pixel dimensions and known scale: 2 cm in image = 50 feet actual.frame5.jpg 

• The circle's diameter was estimated visually based on the image dimensions, and this scale was used to convert pixel measurements into real-world feet. 

• The area was then calculated using the formula for the area of a circle: πr2\pi r^2πr2, with the radius converted to feet.frame5.jpg 

Assumptions 

• The building's central circular feature fully encompasses the occupied area (overestimating the whole structure if non-circular parts extend the actual footprint). 

• Image DPI was assumed to be 96 if metadata was missing, which is standard for web images.frame5.jpg 

• Manual visual estimation introduces possible error, but provides a close real-world approximation given the described scale and visible features. 

• frame5.jpg 

Case 2: 

Prompt:  

Given the bounding box locations of the nearest (Latitude 42.37043, Longitude: -71.12165) and the farthest (Latitude: 42.37125, Longitude: -71.11733) points on the rectangular area covered by the drone in its tour and a scale resolution of 2cm on image = 50 feet on ground, what was the total area in square miles surveyed by the drone. 

Response: 

Ground area covered: 0.012494 square miles 

This can be confirmed with: 

from math import radians, cos, sin, asin, sqrt 
 
# Step 1: Haversine function to compute distances in meters 
def haversine(lat1, lon1, lat2, lon2): 
    R = 6371000  # Earth's radius in meters 
    dlat = radians(lat2 - lat1) 
    dlon = radians(lon2 - lon1) 
    a = sin(dlat/2)**2 + cos(radians(lat1))*cos(radians(lat2))*sin(dlon/2)**2 
    c = 2 * asin(sqrt(a)) 
    return R * c 
 
# Bounding rectangle corners (nearest and farthest) 
lat_min, lon_min = 42.37043, -71.12165 
lat_max, lon_max = 42.37125, -71.11733 
 
# Compute east-west (width) and north-south (height) ground distances, in meters 
height_m = haversine(lat_min, lon_min, lat_max, lon_min) 
width_m = haversine(lat_min, lon_min, lat_min, lon_max) 
 
# Step 2: Area in square meters 
area_m2 = width_m * height_m 
 
# Step 3: Convert to square feet (1 m = 3.28084 ft) 
area_ft2 = area_m2 * (3.28084 ** 2) 
 
# Step 4: Convert to square miles (1 sq mile = 27,878,400 sq ft) 
area_miles2 = area_ft2 / 27878400 
 
print(f"Ground area covered: {area_miles2:.6f} square miles")