Zhiguo Wu

Pyramid architecture is a useful strategy to fuse multi-scale features in deep monocular depth estimation approaches. However, most pyramid networks fuse features only within the adjacent stages in a pyramid structure. To take full advantage of the pyramid structure, inspired by the success of DenseNet, this paper presents DCPNet, a densely connected pyramid network that fuses multi-scale features from multiple stages of the pyramid structure. DCPNet not only performs feature fusion between the adjacent stages, but also non-adjacent stages. To fuse these features, we design a simple and effective dense connection module (DCM). In addition, we offer a new consideration of the common upscale operation in our approach. We believe DCPNet offers a more efficient way to fuse features from multiple scales in a pyramid-like network. We perform extensive experiments using both outdoor and indoor benchmark datasets (i.e., the KITTI and the NYU Depth V2 datasets) and DCPNet achieves the state-of-the-art results.

Pollution of waters resulting from phosphorus (P) runoff from agricultural land receiving long-term manure application is one of the most serious threats to water quality in many regions of the world. Of various approaches to alleviate the problem, reducing P surplus on animal farms through optimizing P intake and minimizing P excretion in manure offers a great opportunity. Here, we present a fecal P test method that has the potential to identify over-feeding of P in dairy cattle. Previous research has suggested that water-extractable P in dairy cow feces closely reflects dietary P changes and may indicate the animal's P status (adequate vs excessive). However, the notion was somewhat confounded when a subsequent study found other factors (pH and Ca content as well as sample handling method) also affecting P extractability in water. In the present work, we hypothesize that the impact of those factors on P extractability can be overcome by selecting dilute acid solutions to replace deionized water as the extractant. Using samples from 25 commercial dairy farms, we tested an array of acid solutions (including HCl, citric acid, and acetic acid) and found that 0.1% HCl is the most suitable extractant. Inorganic P (P_i) released in 0.1% HCl closely reflected dietary P changes among the farms (R² = 0.69) and was independent of pH, Ca, or sample handling method. Knowledge of P metabolism and partitioning in dairy cows and our experimental data suggest that excess P intake by the animal leads to greater amounts of bioavailable but unabsorbed P, which is excreted in feces. Its relative magnitude may be estimated by measuring P_i extractable in 0.1% HCl. This novel and simple fecal P test could potentially be used as an indicator of the animal's P supply utilization status and thus serve as a screening tool for the presence of P over-feeding on dairy farms.