Multi-modal Robustness Analysis Against Language and Visual Perturbations

Anonymous NIPs Submission
[Home]
[Dataset]
[Examples]

Download

To download sample of videos from MSRVTT: [Download Samples]

To download code to generate the benchmarks: [See Code]

Original


Freeze (Temporal)


Box Jumble (Temporal)


Reverse (Temporal)


Impulse (Noise)


Rotate (Camera)


Motion Blur (Blur)