Multiple aspects of the built and natural environment in cities possibly related to people’s walking behavior. Recent transport studies are typically based on density, land use diversity, and urban design to capture the built environmental characteristics. However, streetscape environments capturing how people perceived their surroundings on site are understudied. To respond to this knowledge gap, this study examined possibly non-linear associations between multiple environmental features derived from street view images and peoples’ walking duration in the city of Amsterdam, The Netherlands. We used pooled travel survey data from 2014 to 2017. Streetscape-based environmental measures (e.g., cars and people) were extracted from street view images through a fully convolutional neural network. Covariate-adjusted generalized additive mixed models were fitted to the data. Our results showed that walking-streetscape associations differed between weekdays and weekends. On weekdays, pedestrians walked more in neighborhoods with fewer individual standing walls, lower address density, and poor train accessibility. On weekends, pedestrians’ walking duration increased with more street greenery, fewer cars, higher address density, pronounced land-use diversity, and further distances to train stations. Non-linear associations were found only in the case of weekday street view-derived people, even after adjusting for other neighborhood characteristics (e.g., address density, land use mix, and street connectivity). Our findings suggest that streetscape environmental features complement the typically used built environmental measures to explain pedestrians’ mobility.