Research Topics of Image & Video Coding Group

Multi-Frame Optimized Quantization

We have developed an algorithm which exploits inter-frame dependencies during the quantization stage, such that the overall coding efficiency is improved.

Read more

Perceptually Optimized Video Coding

We evaluate and develop subjective quality models based on the human visual system for the application to image and video coding.

Read more

3D Video Coding

We have developed a comprehensive 3D video coding solution for multiple views and associated depth data, including advanced approaches for inter-view and inter-component prediction.

Read more

3D Video Processing

We have developed 3D video processing technology related to 3D coding, including 3D computer graphics models and free viewpoint video.

Read more

Cross-Component Prediction in HEVC Range Extensions

We have developed the CCP that is a concrete form of the in-loop luma-to-chroma prediction approach exploiting existing and remaining correlation among the color components.

Read more

Generic Quadtree-Based Block Partitioning in HEVC

We have developed a generic quadtree-based block partitioning approach for HEVC that allows a highly flexible adaptation to the signal characteristic.

Read more

Transform Coding Using the Residual Quadtree (RQT)

Transform Coding Using the Residual Quadtree (RQT).

Read more

Interpolation for Motion Compensated Prediction

We have developed approaches for improving motion compensated prediction, including a spline based framework for generating fractional displacements using interpolation filters.

Read more

Block Merging for Quadtree-Based Block Partitioning

Block Merging for Quadtree-Based Block Partitioning.

Read more

Transform Coefficient Level Coding in HEVC

We have developed a transform coefficient level coding scheme for HEVC that improves the compression efficiency and reduces the complexity by a sophisticated design supporting variable transform block sizes.

Read more

Tools for Parallel Processing and Ultra-Low Delay Support

We have developed high-level syntax structures that are specifically designed for improving the parallelization processing capabilities in the implementation of video coding standards.

Read more

Probability Interval Partitioning Entropy (PIPE) Coding

We have developed the probability interval partitioning entropy (PIPE) coding concept. It uses the same flexible and efficient binarization, context modeling, and probability estimation techniques as CABAC while the M coder is replaced with a number of fixed probability bin encoders.

Read more

Context-Based Adaptive Binary Arithmetic Coding (CABAC)

We have developed the context-based adaptive binary arithmetic coding concept. It was originally designed as a flexible and efficient entropy coding scheme for the video compression standard H.264/AVC and is also an integral part of its successor H.265/HEVC.

Read more

Fast Adaptive Binary Arithmetic Coding (M Coder)

We have developed a fast adaptive binary arithmetic coding scheme, which is known as the M coder. It replaces the computational critical multiplication in the interval subdivision of binary arithmetic coding with a table lookup.

Read more

Multi-View Video Coding

We have developed a simple, but very efficient approach for multiview video coding, using inter-view reference pictures for disparity-compensated prediction.

Read more

Rate-Distortion Optimized (RDO) Encoder Control

We are very active in optimizing the encoder control for different video coding standards, the Lagrangian optimization approach has become an essential part of both the H.264/MPEG4-AVC and the HEVC reference encoder software.

Read more

Scalable Video Coding in H.264/AVC

We are investigating generic formats for 3D television and 3D video applications to support a wide range of stereoscopic and autostereoscopic multi-view displays with a single format.

Read more

Inter-Layer Prediction for Scalable Video Coding

For the standardization of SVC, the scalable extension of H.264/MPEG4-AVC, we have developed coding tools which help improving the coding performance by exploiting inter-layer dependencies.

Read more

Hierarchical Prediction Structures

We have investigated and proposed so-called hierarchical prediction structures, which provide both improved coding efficiency and temporal scalability.

Read more

Multi-Frame Motion-Compensated Prediction

We have investigated video coding with multiple reference frames which improves the coding efficiency and therefore has become an essential part of state-of-the-art video coding standards.

Read more