Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.

Relevance-Based Compression of Cataract Surgery Videos Using Convolutional Neural Networks

Recorded cataract surgery videos play a prominent role in training and investigating the surgery, and enhancing the surgical outcomes. Due to storage limitations in hospitals, however, the recorded cataract surgeries are deleted after a short time and this precious source of information cannot be fully utilized. Lowering the quality to reduce the required storage space is not advisable since the degraded visual quality results in the loss of relevant information that limits the usage of these videos. To address this problem, we propose a relevance-based compression technique consisting of two modules: (i) relevance detection, which uses neural networks for semantic segmentation and classification of the videos to detect relevant spatio-temporal information, and (ii) content-adaptive compression, which restricts the amount of distortion applied to the relevant content while allocating less bitrate to irrelevant content. The proposed relevance-based compression framework is implemented considering five scenarios based on the definition of relevant information from the target audience’s perspective. Experimental results demonstrate the capability of the proposed approach in relevance detection. We further show that the proposed approach can achieve high compression efficiency by abstracting substantial redundant information while retaining the high quality of the relevant content.

Related Books

Free with a 30 day trial from Scribd

See all
  • Be the first to comment

  • Be the first to like this

Relevance-Based Compression of Cataract Surgery Videos Using Convolutional Neural Networks

  1. 1. Relevance-Based Compression of Cataract Surgery Videos Using Convolutional Neural Networks Negin Ghamsarian, Hadi Amirpour, Christian Timmerer, Mario Taschwer, Klaus Schöffmann
  2. 2. 1. Introduction 4. Compression Results 2. Proposed Method 3. Detection Results
  3. 3. 1. Introduction:
  4. 4. 1. Introduction:
  5. 5. 1. Introduction 4. Compression Results 2. Proposed Method 3. Detection Results
  6. 6. 3. Proposed Method:
  7. 7. 2. Proposed Method: 1 • 𝐴𝑐𝑡𝑖𝑜𝑛 2 • 𝐴𝑐𝑡𝑖𝑜𝑛 ∩ (𝑐𝑜𝑟𝑛𝑒𝑎 ∪ 𝑖𝑛𝑠𝑡𝑟𝑢𝑚𝑒𝑛𝑡) 3 • 𝐴𝑐𝑡𝑖𝑜𝑛 ∩ 𝑐𝑜𝑟𝑛𝑒𝑎 (simple) 4 • 𝐴𝑐𝑡𝑖𝑜𝑛 ∩ 𝑐𝑜𝑟𝑛𝑒𝑎 (Luma preference) 5 • 𝐴𝑐𝑡𝑖𝑜𝑛 ∩ 𝑐𝑜𝑟𝑛𝑒𝑎 (removed background) Scenarios for relevant content … … … Action Idle Action Idle Irrelevant Irrelevant Relevant
  8. 8. 2. Proposed Method:
  9. 9. 2. Proposed Method:
  10. 10. 2. Proposed Method: Mask R-CNN (Mask Regional Convolutional Neural Network) Two sub-problems: Object Detection Semantic Segmentation Using region proposal networks Using fully convolutional networks
  11. 11. 2. Proposed Method: 9 proposals relative to 9 anchors, and 4 parameters per anchor Features from the backbone network 9 × 4 = 36 Relative box parameters Class probabilities 9 × 2 = 18 RPN (Region Proposal Network):
  12. 12. 2. Proposed Method: RPN (Region Proposal Network): 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1 2 3 4 5 6 7 8 9 Class Probabilities Cornea Background 1 2 3 4 75 6 8 9
  13. 13. 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1 2 3 4 5 6 7 8 9 Class Probabilities Cornea Background 3. Proposed Method: RPN (Region Proposal Network): 1 2 3 4 75 6 8 9
  14. 14. 2. Proposed Method: Forward/Inference Backward/Learning FCN (Fully Convolutional Network):
  15. 15. 1. Introduction 4. Compression Results 2. Proposed Method 3. Detection Results
  16. 16. 3. Detection Results:
  17. 17. 3. Detection Results: V2 V5 V7 V8 Idle Action
  18. 18. 3. Detection Results:
  19. 19. 3. Detection Results: At least 80% IoU for mask segmentation
  20. 20. 3. Detection Results: At least 85% IoU for bounding-box segmentation
  21. 21. 1. Introduction 4. Compression Results 2. Proposed Method 3. Detection Results
  22. 22. 5. Compression Results: ∆𝑄 = 5 ∆𝑄 = 10 ∆𝑄 = 13 ∆𝑄 = 15 PSNR values for an exemplary segment using different QP differences:
  23. 23. 5. Compression Results: The percentage of bitrate reduction resulting from different scenarios and different QP differences:
  24. 24. 5. Compression Results: The PSNR of ROI and output size corresponding to an exemplary video compressed in different scenarios:
  25. 25. 5. Compression Results: Original Scenario II
  26. 26. 5. Compression Results: Original Scenario IV
  27. 27. 5. Compression Results: Original Scenario V
  28. 28. Conclusion: 4 Being generalizable to all sorts of surgical videos where the domain knowledge is required 3 A set of novel scenarios to compress cataract surgery videos so as to be proportionate to the target audience.2 1 Main Contributions Relevance-based compression approach using surgical domain knowledge Storage space gain of up to 68%
  29. 29. Question? negin@itec.aau.at hadi@itec.aau.at

×