3D-Image Rendering TRIZ Cases
- Text-Guided 3D Model View Generation with Contour Control
Combines a 3D geometric model with text to add texture while preserving contour similarity and enabling interactive view previews.
- Synthetic 3D X-Ray Bag Images for Threat Detection Training
Scarce and risky real-world scans are replaced by combined bag and threat volumes that expand 3D CT detection training data.
- Ray Tracing Reservoir Resampling for Dynamic Global Illumination
Temporal and spatial reservoir buffers resample illumination samples to reduce ray-tracing compute and memory demands in dynamic scenes.
- Wide-Field Image Screen Generation With 3D Position Linking
Position-linked display pairs a planar image area with 3D content, making wide-field and 360-degree scenes easier to view.
- VR Volumetric Video Editing for Spatiotemporal Consistency
VR users can add, remove, or reposition volumetric-video objects while AI and generative models preserve spatiotemporal consistency.
- Omnidirectional 3D Gaussian Rendering With Polar Stretching
Polar stretching adapts 3D Gaussian rendering to omnidirectional camera distortion while preserving real-time parallel processing.
- 3D Implicit Neural Representation Training With Geometric Priors
Laplacian features and a denoising neural network guide implicit representations to accelerate convergence and reduce 3D reconstruction training time.
- Synthetic Depth Image Pairs for Missing-Value Restoration
Synthetic scenes simulate sensor shadows and absorbing objects to train models that restore missing depth values in real images.
- Eye-Tracking Responsiveness Adjustment for Accurate Subject Focus
A gaze-tracking apparatus compares eye and object movement to adjust responsiveness, suppress involuntary motion, and improve intended-subject focus.
- Hybrid Rendering With Deferred Primitive Binning to Reduce Shading Overhead
Repeated color and depth reads burden immediate-mode rendering; deferred primitive batches discard non-contributing fragments before shading.