This paper proposes an efficient multiresolution motion estimation algorithm (MMEA) well-suited for high definition video encoder VLSI implementation. 16-way parallel process element arrays achieve fast search with regular data access and efficient data reuse. Moreover, the parallel arrays can be efficiently reused at three hierarchical levels for sequential motion vector (MV) refinement. Also, an intelligent algorithm for multiple candidate MV selection is proposed to compensate for performance degradation due to parallel MMEA. The whole algorithm achieves a good balance between complexity and performance. Simulation results show that only 0.14dB PSNR degradation at the worst case is achieved compared with full search algorithm. ? 2009 IEE...
This thesis deals with the design and implementation of extremely parallel fast motion / disparity e...
Motion estimation is the most complex module which contributes nearly 70% of computation resources i...
This paper presents a paralleled Two-Pass Hexagonal (TPA) algorithm constituted by Linear Hashtable ...
This paper proposes a hardware friendly multi-resolution motion estimation algorithm and VLSI archit...
High throughput, heavy bandwidth requirement, huge on-chip memory consumption, and complex data flow...
Abstract—High throughput, heavy bandwidth requirement, huge on-chip memory consumption, and complex ...
In this paper, we propose a hardware architecture of an adaptive multi-resolution motion estimation ...
In this paper, we propose a hardware architecture of an adaptive multi-resolution motion estimation ...
Integer motion estimation (IME) for block-based video coding presents a significant challenge in ext...
The paper presents a hardware-efficient fast algorithm and its architecture for large search range m...
We propose a highly parallel and scalable motion estimation algorithm, named multilevel resolution m...
This paper presents a VLSI architecture for a low complexity motion estimation algorithm, referred t...
In this paper, we present a fast multi-resolution block matching algorithm (BMA) for an MPEG-2 video...
15 pagesInternational audienceThe effciency of video compression methods mainly depends on the motio...
Motion Estimation (ME) is the most computationally intensive part in the whole video compression pro...
This thesis deals with the design and implementation of extremely parallel fast motion / disparity e...
Motion estimation is the most complex module which contributes nearly 70% of computation resources i...
This paper presents a paralleled Two-Pass Hexagonal (TPA) algorithm constituted by Linear Hashtable ...
This paper proposes a hardware friendly multi-resolution motion estimation algorithm and VLSI archit...
High throughput, heavy bandwidth requirement, huge on-chip memory consumption, and complex data flow...
Abstract—High throughput, heavy bandwidth requirement, huge on-chip memory consumption, and complex ...
In this paper, we propose a hardware architecture of an adaptive multi-resolution motion estimation ...
In this paper, we propose a hardware architecture of an adaptive multi-resolution motion estimation ...
Integer motion estimation (IME) for block-based video coding presents a significant challenge in ext...
The paper presents a hardware-efficient fast algorithm and its architecture for large search range m...
We propose a highly parallel and scalable motion estimation algorithm, named multilevel resolution m...
This paper presents a VLSI architecture for a low complexity motion estimation algorithm, referred t...
In this paper, we present a fast multi-resolution block matching algorithm (BMA) for an MPEG-2 video...
15 pagesInternational audienceThe effciency of video compression methods mainly depends on the motio...
Motion Estimation (ME) is the most computationally intensive part in the whole video compression pro...
This thesis deals with the design and implementation of extremely parallel fast motion / disparity e...
Motion estimation is the most complex module which contributes nearly 70% of computation resources i...
This paper presents a paralleled Two-Pass Hexagonal (TPA) algorithm constituted by Linear Hashtable ...