#object detection

[alphatree]

Aircraft Detection on VHR imagery

Results of aircraft detection (accuracy of 100%) © Atlas AI, on varying resolution VHR imagery (Maxar, Satellogic, Airbus). Detecting and classifying aircrafts allows us to better understand the commercial activity in key locations around the world in support of our socio-economic analyses.

The method features advanced deep learning technologies, and model training uses data acquired from a rich pool of diverse imagery, all normalized to uniform standards using proprietary algorithmics.

[prostheticknowledge]

YOLO v3

New release of Object Recognition framework offers faster and improved realtime performance:

You only look once (YOLO) is a state-of-the-art, real-time object detection system. On a Pascal Titan X it processes images at 30 FPS and has a mAP of 57.9% on COCO test-dev. 

... Prior detection systems repurpose classifiers or localizers to perform detection. They apply the model to an image at multiple locations and scales. High scoring regions of the image are considered detections.

We use a totally different approach. We apply a single neural network to the full image. This network divides the image into regions and predicts bounding boxes and probabilities for each region. These bounding boxes are weighted by the predicted probabilities.

Our model has several advantages over classifier-based systems. It looks at the whole image at test time so its predictions are informed by global context in the image. It also makes predictions with a single network evaluation unlike systems like R-CNN which require thousands for a single image. This makes it extremely fast, more than 1000x faster than R-CNN and 100x faster than Fast R-CNN. 

More Here

[algopop]

This is the result of an overly sensitive object detection algorithm, it is overwhelmed by everything it sees. 

[tangibledisplay]

We returned to Retail Design Expo 2016 with a new interactive installation exploring creative collaboration. The concept centres on interfaces activated and controlled by the audience. Visitors are invited to place a selection of everyday objects – a box, a plant, a smartphone, a coffee pot – from a shelf onto any of four sensors to trigger bespoke film content and accompanying soundtrack layers. Each film has been created by a different designer, offering their individual response to a specific object for an insight into our various disciplines and the people working within them. The markers correspond to our four main areas of expertise – digital, graphics, interiors and strategy – and the content for each object changes to reflect this, triggered wirelessly by embedded radio-frequency identification (RFID) tags. Though the technology itself is simple, the installation’s impact comes from the wealth of original content and surprising interactions. Some films are comical, others more serious meditations on the creative process. Together, they illustrate how we can create character and compelling storytelling from any given starting point. By playing with different sequences of objects, participants can exercise their own creativity by ‘remixing’ the visuals and audio, creating layered montages that explore the agency’s people, processes and creative spirit.

[prostheticknowledge]

CASENet: Deep Category-Aware Semantic Edge Detection

Computer Vision paper from Mitsubishi Electric Research Labs (MERL) uses machine learning to break down a visual feed into identifiable parts:

Boundary and edge cues are highly beneficial in improving a wide variety of vision tasks such as semantic segmentation, object recognition, stereo, and object proposal generation. Recently, the problem of edge detection has been revisited and significant progress has been made with deep learning. While classical edge detection is a challenging binary problem in itself, the category-aware semantic edge detection by nature is an even more challenging multi-label problem. We model the problem such that each edge pixel can be associated with more than one class as they appear in contours or junctions belonging to two or more semantic classes. To this end, we propose a novel end-to-end deep semantic edge learning architecture based on ResNet and a new skip-layer architecture where category-wise edge activations at the top convolution layer share and are fused with the same set of bottom layer features. We then propose a multi-label loss function to supervise the fused activations. We show that our proposed architecture benefits this problem with better performance, and we outperform the current state-of-the-art semantic edge detection methods by a large margin on standard data sets such as SBD and Cityscapes. 

Link

[algopop]

An algorithm watching a movie trailer  by Støj

Everything but the detected objects are removed from the trailer of The Wolf of Wall Street. The software is powered by Yolo object-detection, which has been used for similar experiments. 

[marwizteach]

Object Loading & Unloading Detection System | Marwiz Vision

Smart object detection for real-time loading and unloading monitoring, improving warehouse efficiency, security, and accountability.

[liki281]

YOLO V/s Embeddings: A comparison between two object detection models

YOLO-Based Detection Model Type: Object detection Method: YOLO is a single-stage object detection model that divides the image into a grid and predicts bounding boxes, class labels, and confidence scores in a single pass. Output: Bounding boxes with class labels and confidence scores. Use Case: Ideal for real-time applications like autonomous vehicles, surveillance, and robotics. Example Models: YOLOv3, YOLOv4, YOLOv5, YOLOv8 Architecture

YOLO processes an image in a single forward pass of a CNN. The image is divided into a grid of cells (e.g., 13×13 for YOLOv3 at 416×416 resolution). Each cell predicts bounding boxes, class labels, and confidence scores. Uses anchor boxes to handle different object sizes. Outputs a tensor of shape [S, S, B*(5+C)] where: S = grid size (e.g., 13×13) B = number of anchor boxes per grid cell C = number of object classes 5 = (x, y, w, h, confidence) Training Process

Loss Function: Combination of localization loss (bounding box regression), confidence loss, and classification loss.

Labels: Requires annotated datasets with labeled bounding boxes (e.g., COCO, Pascal VOC).

Optimization: Typically uses SGD or Adam with a backbone CNN like CSPDarknet (in YOLOv4/v5). Inference Process

Input image is resized (e.g., 416×416). A single forward pass through the model. Non-Maximum Suppression (NMS) filters overlapping bounding boxes. Outputs detected objects with bounding boxes. Strengths

Fast inference due to a single forward pass. Works well for real-time applications (e.g., autonomous driving, security cameras). Good performance on standard object detection datasets. Weaknesses

Struggles with overlapping objects (compared to two-stage models like Faster R-CNN). Fixed number of anchor boxes may not generalize well to all object sizes. Needs retraining for new classes. Embeddings-Based Detection Model Type: Feature-based detection Method: Instead of directly predicting bounding boxes, embeddings-based models generate a high-dimensional representation (embedding vector) for objects or regions in an image. These embeddings are then compared against stored embeddings to identify objects. Output: A similarity score (e.g., cosine similarity) that determines if an object matches a known category. Use Case: Often used for tasks like face recognition (e.g., FaceNet, ArcFace), anomaly detection, object re-identification, and retrieval-based detection where object categories might not be predefined. Example Models: FaceNet, DeepSORT (for object tracking), CLIP (image-text matching) Architecture

Uses a deep feature extraction model (e.g., ResNet, EfficientNet, Vision Transformers). Instead of directly predicting bounding boxes, it generates a high-dimensional feature vector (embedding) for each object or image. The embeddings are stored in a vector database or compared using similarity metrics. Training Process Uses contrastive learning or metric learning. Common loss functions:

Triplet Loss: Forces similar objects to be closer and different objects to be farther in embedding space.

Cosine Similarity Loss: Maximizes similarity between identical objects.

Center Loss: Ensures class centers are well-separated. Training datasets can be either:

Labeled (e.g., with identity labels for face recognition).

Self-supervised (e.g., CLIP uses image-text pairs). Inference Process

Extract embeddings from a new image using a CNN or transformer. Compare embeddings with stored vectors using cosine similarity or Euclidean distance. If similarity is above a threshold, the object is recognized. Strengths

Scalable: New objects can be added without retraining.

Better for recognition tasks: Works well for face recognition, product matching, anomaly detection.

Works without predefined classes (zero-shot learning). Weaknesses

Requires a reference database of embeddings. Not good for real-time object detection (doesn’t predict bounding boxes directly). Can struggle with hard negatives (objects that look similar but are different).

Weaknesses

Struggles with overlapping objects (compared to two-stage models like Faster R-CNN). Fixed number of anchor boxes may not generalize well to all object sizes. Needs retraining for new classes. Embeddings-Based Detection Model Type: Feature-based detection Method: Instead of directly predicting bounding boxes, embeddings-based models generate a high- dimensional representation (embedding vector) for objects or regions in an image. These embeddings are then compared against stored embeddings to identify objects. Output: A similarity score (e.g., cosine similarity) that determines if an object matches a known category. Use Case: Often used for tasks like face recognition (e.g., FaceNet, ArcFace), anomaly detection, object re-identification, and retrieval-based detection where object categories might not be predefined. Example Models: FaceNet, DeepSORT (for object tracking), CLIP (image-text matching) Architecture

Uses a deep feature extraction model (e.g., ResNet, EfficientNet, Vision Transformers). Instead of directly predicting bounding boxes, it generates a high-dimensional feature vector (embedding) for each object or image. The embeddings are stored in a vector database or compared using similarity metrics. Training Process Uses contrastive learning or metric learning. Common loss functions:

Triplet Loss: Forces similar objects to be closer and different objects to be farther in embedding space.

Cosine Similarity Loss: Maximizes similarity between identical objects.

Center Loss: Ensures class centers are well-separated. Training datasets can be either:

Labeled (e.g., with identity labels for face recognition).

Self-supervised (e.g., CLIP uses image-text pairs). Inference Process

Extract embeddings from a new image using a CNN or transformer. Compare embeddings with stored vectors using cosine similarity or Euclidean distance. If similarity is above a threshold, the object is recognized. Strengths

Scalable: New objects can be added without retraining.

Better for recognition tasks: Works well for face recognition, product matching, anomaly detection.

Works without predefined classes (zero-shot learning). Weaknesses

Requires a reference database of embeddings. Not good for real-time object detection (doesn’t predict bounding boxes directly). Can struggle with hard negatives (objects that look similar but are different).