Tri-plane Representation and Analysis in Transform Coding
Feifei Zhou, Rui Yang, Bo Li · Advances in intelligent systems research/Advances in Intelligent Systems Research · 2013
Representation and coding of transform coefficients is a key factor in image compression.In this work, we introduce a coefficient representation model, called tri-plane.The proposed method is able to bring more kinds of symbols and significance states, which deeply develops the correlations among transform coefficients.By utilizing new symbols and states, more precise classification can be achieved, which directly determines the prediction coding performance.Experimental results show that tri-plane representation outperforms the traditional bit-plane form in entropy, which removes more redundancy among coefficients and obtains more potential to improve coding efficiency.We hope this work would motivate more research and application work along this direction.