1: \begin{abstract}

2: % Body of abstract:

3: {The problem of recovering a low $n$-rank tensor is an extension of sparse recovery problem from the low dimensional

4: space (matrix space) to the high dimensional space (tensor space) and has many applications in computer vision and graphics

5: such as image inpainting and video inpainting. In this paper, we consider a new tensor recovery model, named as minimum

6: $n$-rank approximation (MnRA), and propose an appropriate iterative hard thresholding algorithm with giving the upper bound

7: of the $n$-rank in advance. The convergence analysis of the proposed algorithm is also presented. Particularly, we show that for the noiseless case, the linear convergence with rate $\frac{1}{2}$ can be obtained for the proposed algorithm under proper conditions.

8: Additionally, combining an effective heuristic for determining $n$-rank, we can also apply the proposed algorithm to solve MnRA when $n$-rank is unknown in advance. Some preliminary numerical results on randomly generated and real low $n$-rank tensor

9: completion problems are reported, which show the efficiency of the proposed algorithms.}

10: % Keywords:

11: {iterative hard thresholding; low-$n$-rank tensor recovery; tensor completion; compressed sensing}

12: \end{abstract}