In the Journals

Initial blurry vision in infants fundamental to normal visual development

Researchers propose that initial poor retinal acuity in infants may have adaptive value.

They used a deep convolutional neural network trained on a large database of facial images and simulated three different visual training scenarios by systematically blurring and increasing the resolution of the images.

In the first visual training scenario, they replicated an infant’s normal course of vision by training the network on blurry images, followed by high-resolution ones. In the second, the training order was reversed, where high-resolution images were followed by low-resolution ones. In the third scenario, the network was trained on only high-resolution images.

The researchers showed that commencing training with blurred images creates receptive fields that integrate information across larger image areas and leads to improved performance and better generalization across a range of resolutions, according to the study.

Children who never experience the initial phase of out-of-focus imagery can later have difficulty with complex visual tasks, like those who are born blind from cataracts and undergo corrective surgery later in childhood, according to an NEI press release about the study.

These patients have been observed to have long-term difficulty with facial recognition, according to Pawan Sinha, PhD, senior study investigator, in the release.

“The rapid attainment of relatively high visual acuity after cataract surgery may be an important factor that pushes a child’s developmental trajectory away from the normal one, wherein acuity improves gradually after initially being quite poor,” he said.

“From a practical perspective, these findings suggest a superior strategy for training deep neural networks,” Sinha added. “Training neural networks on high resolution images has been a common default. Perhaps image recognition could be improved by drawing inspiration from human development, and training networks first with degraded images.”

The findings also suggest that children who have surgery for cataracts may benefit from a more gradual improvement in their visual acuity, according to the release.

“These findings offer an explanation for the observed face recognition impairments after late treatment of congenital blindness, suggest an adaptive function for the acuity trajectory in normal development and provide a scheme for improving the performance of computational face recognition systems,” researchers wrote. – by Abigail Sutton

Disclosures: The researchers reported no relevant financial disclosures.

Researchers propose that initial poor retinal acuity in infants may have adaptive value.

They used a deep convolutional neural network trained on a large database of facial images and simulated three different visual training scenarios by systematically blurring and increasing the resolution of the images.

In the first visual training scenario, they replicated an infant’s normal course of vision by training the network on blurry images, followed by high-resolution ones. In the second, the training order was reversed, where high-resolution images were followed by low-resolution ones. In the third scenario, the network was trained on only high-resolution images.

The researchers showed that commencing training with blurred images creates receptive fields that integrate information across larger image areas and leads to improved performance and better generalization across a range of resolutions, according to the study.

Children who never experience the initial phase of out-of-focus imagery can later have difficulty with complex visual tasks, like those who are born blind from cataracts and undergo corrective surgery later in childhood, according to an NEI press release about the study.

These patients have been observed to have long-term difficulty with facial recognition, according to Pawan Sinha, PhD, senior study investigator, in the release.

“The rapid attainment of relatively high visual acuity after cataract surgery may be an important factor that pushes a child’s developmental trajectory away from the normal one, wherein acuity improves gradually after initially being quite poor,” he said.

“From a practical perspective, these findings suggest a superior strategy for training deep neural networks,” Sinha added. “Training neural networks on high resolution images has been a common default. Perhaps image recognition could be improved by drawing inspiration from human development, and training networks first with degraded images.”

The findings also suggest that children who have surgery for cataracts may benefit from a more gradual improvement in their visual acuity, according to the release.

“These findings offer an explanation for the observed face recognition impairments after late treatment of congenital blindness, suggest an adaptive function for the acuity trajectory in normal development and provide a scheme for improving the performance of computational face recognition systems,” researchers wrote. – by Abigail Sutton

Disclosures: The researchers reported no relevant financial disclosures.