Dimensions overall score 3.0

GitHub Code Pulse

No public code linked for this paper yet.

• Object detection is performed by identifying all high-scoring patches across all possible scales and overlapping them to form a bounding box.method 90%
• Training DisCNN requires only object-centered image data with positive and negative class labels.method 90%
• The modules of the output vector of a DisCNN with respect to a specific positive class are positively monotonic with the presence probabilities of the positive features.technical 90%
• Detection for multiple positive classes can be conducted in parallel to significantly accelerate the process.technical 85%
• A DisCNN for one positive class is lightweight, which reduces the time of both training and detection.technical 85%
• The DisCNN architecture is based on VGG but reduces the number of feature maps to 8 and omits the softmax layer.method 95%
• The DisCNN is trained using a novel loss function called Negatives to Origin (N2O).method 95%
• The method enables the detection of sub-features of an object.result 80%

Competitive landscape

Competitor map is still being generated for this paper. Enable generation or check back soon.

Startup potential card