Enhancing Deep Neural Networks for Real-Time Image Classification: A Comparative Analysis of Optimization Techniques
Bibliographic record
Abstract
The use of Deep Neural Networks (DNNs) for picture classification has been very successful in many different industries. Computational complexity, latency concerns, and the need for great efficiency make their use in real-time applications difficult. A study that compares optimization methods with the goal of making DNNs better at classifying images in real-time. We assess various approaches, such as weight pruning, quantization, low-rank factorization, and knowledge distillation, taking into consideration their effects on model precision, inference velocity, and computing demands. We use state-of-the-art DNN architectures like ResNet and MobileNet to gain experimental results from popular picture datasets like CIFAR-10 and ImageNet. Our research shows that although model efficiency and accuracy are not always compatible, that pruning and quantization are two optimization methods that can greatly reduce inference time while keeping classification accuracy relatively stable. When it comes to selecting the right optimization strategies for deploying DNNs in real-time, mission-critical applications like autonomous driving, video surveillance, and augmented reality systems, we also investigate hybrid approaches that combine various optimizations to further decrease latency and improve performance in environments with limited resources.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".