Imaging cerebral blood flow (CBF) is crucial for diagnosing cerebrovascular diseases. Conventional laser speckle contrast imaging (LSCI) offers high spatiotemporal resolution but limited penetration depth (<1 mm). This study introduces an innovative time-resolved laser speckle contrast imaging (TR-LSCI) technique, which illuminates picosecond-pulsed, coherent, widefield, near-infrared light onto the head and synchronizes a newly launched, ultrafast, high-resolution, picosecond-gated SPAD512^2 camera to continuously and rapidly capture CBF maps at different depths, thus reducing partial volume effects from the overlying skull. The new-generation TR-LSCI enabled 2D mapping of pulsatile CBF distributions in rodents with a fast-sampling rate of 21.1 Hz. The FFT analyses of pulsatile CBF maps revealed multiple harmonic frequency peaks, which corresponded to respiratory and heartbeat signals. The new TR-LSCI with high spatiotemporal resolution captures fast hemodynamic changes across different brain regions and generates large spatiotemporal datasets for extracting intracranial pressure information from measured pulsatile CBF waveforms using deep learning algorithms.