Inventor

PORIKLI FATIH MURAT

US41 patents

⚠️ This page may combine multiple inventors who share the name “PORIKLI FATIH MURAT”. Patents are grouped by organization below to help distinguish them — per-person disambiguation is on the roadmap.

PORIKLI FATIH MURAT

QUALCOMM INC

Volumetric sampling with correlative characterization for dense estimation

Gradient pruning for efficient training of machine learning models

Model disentanglement for domain adaptation

Desparsified convolution for sparse activations

Efficient neural-network-based processing of visual content

Sparse optical flow estimation

Volumetric sampling with correlative characterization for dense estimation

Occlusion-aware forward warping for video frame interpolation

Domain adaptation for wireless sensing

Zero-shot deep learning for multi-target RF positioning

Efficient test-time adaptation for improved temporal consistency in video processing

Supervised learning and occlusion masking for optical flow estimation

Self-supervised passive positioning using wireless data

Dimensionality transformation for efficient bottleneck processing

Saliency-based input resampling for efficient object detection

Unsupervised object-oriented decompositional normalizing flow

Efficient pose estimation through iterative refinement

Sparse-aware convolution and techniques for acceleration thereof

Efficient video processing via dynamic knowledge propagation

Multi-modal beam prediction

Parameterized activation functions to adjust model linearity

Consistency measure for image segmentation processes

Spatiotemporal recycling network

Online adaptation of segmentation machine learning systems

Virtual representative conditioning system

Three-dimensional object part segmentation using a machine learning model

Recurrent neural networks for high resolution image and video processing

Panoptic segmentation with panoptic, instance, and semantic relations

Apparatus and methods for image segmentation using machine learning processes

Data efficient learning and rapid domain adaptation for wireless positioning and tracking

Adaptive human instance segmentation with stereo view consistency

Unified simultaneous optical flow and depth estimation

Planar mesh reconstruction using images from multiple camera poses

Disparity-based depth refinement using confidence information and stereoscopic depth information

Text-guided image editing by learning guidance scales via reinforcement learning

Adaptive personalization for anti-spoofing protection in biometric authentication systems

Depth map completion in visual content using semantic and three-dimensional information

Meta-pre-training with augmentations to generalize neural network processing for domain adaptation

Neural network channel-wise tiling with overlaps

SHENZHEN YINWANG INTELLIGENT TECHNOLOGY CO LTD

Driver attention detection method

HUAWEI TECH CO LTD

Multi-stage multi-reference bootstrapping for video super-resolution