Multiple-input multiple-output orthogonal frequency division multiplexing (MIMO-OFDM) technology is a promising solution for next generation wireless communications, due to high bandwidth efficiency, resistance to RF interference, and robustness to multipath fading. A major drawback of OFDM is its high peak-to-average power ratio (PAPR) which results in non-linearities in the output signal. In this paper, two methods based on spatial/temporal processing are proposed to reduce the PAPR of MIMO-OFDM systems. These methods divide the OFDM block at each transmit antenna into some subblocks. Then, spatial and temporal processing in the form of circular shifting or interleaving are applied to generate different candidate sequences. Finally, for each transmit antenna the candidate sequence with the lowest PAPR is chosen for transmission. Compared to the conventional PAPR reduction schemes such as ordinary partial transmit sequences (O-PTS), the proposed methods require lower computational complexity and have superior PAPR reduction performance.