There is a lack of data on physical activity (PA), active travel, and the comparison of measurement instruments in low-resource settings. The objective of this paper is to describe PA behaviour and the agreement of walking estimates from the Global Physical Activity Questionnaire (GPAQ) and the travel diary in a low-resource setting. We used a cross-sectional survey design to capture data from the residents of Accra (Ghana) between May 2020 and March 2021. Of the 863 participants aged 15+ years, 65% were females, and 86% reported PA. The median weekly PA was 18 (interquartile range: 5–75) metabolic equivalent of task hours, with 50% of females and 37% of males achieving low PA levels. In the GPAQ, 80% of participants reported weekly walking; the mean number of days walked was 3.8 (standard deviation (SD): 2.5); hence, 54% of participants reported walking on any day, and the mean daily walking duration was 51 (SD: 82) minutes. In the diary, 56% of participants reported walking for over 24 h, with a mean walking duration of 31 (SD: 65) minutes. The correlation of walking duration between instruments was weak (rho: 0.31; 95% Confidence Interval: 0.25–0.37); the mean bias was 20 min, with GPAQ estimates being 0.1 to 9 times higher than diary estimates. We concluded that low PA is prevalent in Accra, and while the travel diary and GPAQ estimate similar walking prevalence, their walking duration agreement is poor. We recommend accompanying PA questionnaires with objective measures for calibration.