|Trend||Dataset||Best Method||Paper title||Paper||Code||Compare|
Weight and activation binarization is an effective approach to deep neural network compression and can accelerate the inference by leveraging bitwise operations.
Time-series of satellite images may reveal important data about changes in environmental conditions and natural or urban landscape structures that are of potential interest to citizens, historians, or policymakers.
In 16 bits, our log float multiply-add is 0. 59x the power and 0. 68x the area of IEEE 754 float16 fused multiply-add, maintaining the same signficand precision and dynamic range, proving useful for training ASICs as well.
Armed with this knowledge, we quantize the model parameters, activations and layer gradients to 8-bit, leaving at a higher precision only the final step in the computation of the weight gradients.
A first estimate of the disparity is computed in a very low resolution cost volume, then hierarchically the model re-introduces high-frequency details through a learned upsampling function that uses compact pixel-to-pixel refinement networks.