Advanced search

Found 4 essays matching the above criteria.

1. 1. Deep Learning Based Out-of-focus Detection on Surveillance Footage

   University essay from Lunds universitet/Matematik LTH

   Author : Hanna Bengtsson; Teodora Kerac; [2023]
   Keywords : Deep Learning; CNN; Out-of-focus detection; Mathematics and Statistics;

   Abstract : Loss of focus in surveillance cameras can occur due to factors such as environmental aspects, sabotage or system errors. The focus loss leads to degraded footage that is of no use for either post-investigation or automatic video analysis. READ MORE

3. 2. Resource Efficient Representation of Machine Learning Models : investigating optimization options for decision trees in embedded systems

   University essay from Linköpings universitet/Statistik och maskininlärning

   Author : Jacob Lundberg; [2019]
   Keywords : machine learning; rule fit; decision trees; embedded systems; resources; ensemble methods; lasso; regression; optimization;

   Abstract : Combining embedded systems and machine learning models is an exciting prospect. However, to fully target any embedded system, with the most stringent resource requirements, the models have to be designed with care not to overwhelm it. Decision tree ensembles are targeted in this thesis. READ MORE

4. 3. Investigating the Adaptive Loop Filter in Next Generation Video Coding

   University essay from KTH/Skolan för elektro- och systemteknik (EES)

   Author : Alfonso De La Rocha Gomes-Arevalillo; [2017]
   Keywords : Joint Exploration Model; JEM; Adaptive Loop Filter; variance-based adaptive quantization; video compression; video coding standards;

   Abstract : Current trends on video technologies and services are demanding higher bit rates, highervideo resolutions and better video qualities. This issue results in the need of a new generationof video coding techniques to increase the quality and compression rates of previous standards. READ MORE

5. 4. Compact ConvNets with Ternary Weights and Binary Activations

   University essay from KTH/Robotik, perception och lärande, RPL

   Author : Ondrej Holesovsky; [2017]
   Keywords : convolutional neural networks; compact architectures; weight quantization; activation quantization; dithering activations;

   Abstract : Compact architectures, ternary weights and binary activations are two methods suitable for making neural networks more efficient. We introduce a) a dithering binary activation which improves accuracy of ternary weight networks with binary activations by randomizing quantization error, and b) a method of implementing ternary weight networks with binary activations using binary operations. READ MORE